<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
	>

<channel>
	<title>Amazon Comprehend Medical | John Nessime</title>
	<atom:link href="https://john-nessime.com/blog/tag/amazon-comprehend-medical/feed/" rel="self" type="application/rss+xml" />
	<link>https://john-nessime.com/blog/tag/amazon-comprehend-medical/</link>
	<description>Cloud, DevOps, Data &#38; AI — Built, Tested, Explained</description>
	<lastBuildDate>Mon, 14 Sep 2026 13:13:18 +0000</lastBuildDate>
	<language>en-US</language>
	<sy:updatePeriod>
	hourly	</sy:updatePeriod>
	<sy:updateFrequency>
	1	</sy:updateFrequency>
	<generator>https://wordpress.org/?v=7.1.4</generator>

<image>
	<url>https://john-nessime.com/blog/wp-content/uploads/2026/07/cropped-jn-32x32.png</url>
	<title>Amazon Comprehend Medical | John Nessime</title>
	<link>https://john-nessime.com/blog/tag/amazon-comprehend-medical/</link>
	<width>32</width>
	<height>32</height>
</image> 
	<item>
		<title>Where PHI Escapes: Building a Secure Medical Lab Data Pipeline on AWS</title>
		<link>https://john-nessime.com/blog/cloud-computing/medical-lab-data-pipeline-aws/</link>
		
		<dc:creator><![CDATA[John Nessime]]></dc:creator>
		<pubDate>Fri, 21 Aug 2026 09:00:00 +0000</pubDate>
				<category><![CDATA[Cloud Computing]]></category>
		<category><![CDATA[Cloud Security]]></category>
		<category><![CDATA[Compliance]]></category>
		<category><![CDATA[Data Engineering]]></category>
		<category><![CDATA[Amazon Comprehend Medical]]></category>
		<category><![CDATA[Amazon Macie]]></category>
		<category><![CDATA[Amazon S3]]></category>
		<category><![CDATA[AWS]]></category>
		<category><![CDATA[AWS HealthLake]]></category>
		<category><![CDATA[AWS KMS]]></category>
		<category><![CDATA[Business Associate Agreement]]></category>
		<category><![CDATA[CLIA]]></category>
		<category><![CDATA[Clinical Data]]></category>
		<category><![CDATA[Crypto Shredding]]></category>
		<category><![CDATA[Data Classification]]></category>
		<category><![CDATA[De-identification]]></category>
		<category><![CDATA[Dead Letter Queue]]></category>
		<category><![CDATA[ePHI]]></category>
		<category><![CDATA[FHIR]]></category>
		<category><![CDATA[Healthcare Cloud]]></category>
		<category><![CDATA[HIPAA]]></category>
		<category><![CDATA[HL7 v2]]></category>
		<category><![CDATA[Log Retention]]></category>
		<category><![CDATA[Multi-Account Strategy]]></category>
		<category><![CDATA[PHI]]></category>
		<category><![CDATA[Pipeline Design]]></category>
		<category><![CDATA[Pseudonymization]]></category>
		<category><![CDATA[S3 Object Lock]]></category>
		<guid isPermaLink="false">https://john-nessime.com/blog/?p=455</guid>

					<description><![CDATA[<p>Encryption is the part everyone gets right. The PHI sitting in your logs, dead letter queues and object key names is the part that fails an audit. A boundary-by-boundary guide to building a medical lab data pipeline on AWS that survives review.</p>
<p>The post <a href="https://john-nessime.com/blog/cloud-computing/medical-lab-data-pipeline-aws/">Where PHI Escapes: Building a Secure Medical Lab Data Pipeline on AWS</a> appeared first on <a href="https://john-nessime.com/blog">John Nessime</a>.</p>
]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">The question that breaks most healthcare data designs is not technical. It is an auditor asking you to list every place protected health information is stored. Someone points at the encrypted S3 bucket, the encrypted RDS instance, the encrypted backups, and says &#8220;those three.&#8221; Then you grep CloudWatch Logs and find an accession number, a date of birth, and a patient surname sitting inside a Python traceback with no retention policy set on the log group.</p>



<p class="wp-block-paragraph">That is the failure mode that actually bites when you build a medical lab data pipeline on AWS. Not the bucket you thought about. The seven places PHI ended up because a payload got logged, an error got queued, or a filename contained a medical record number.</p>



<p class="wp-block-paragraph">This post walks through the design boundary by boundary: where lab data crosses from one trust zone into another, what leaks at each crossing, and which control closes it. It is aimed at engineers who have been handed an HL7 feed and a compliance checklist and told to make them meet. I am an engineer and not your compliance counsel, so treat the regulatory references here as pointers to read the primary text yourself.</p>



<h2 class="wp-block-heading">Why laboratory data is a harder pipeline than it looks</h2>



<p class="wp-block-paragraph">Lab data has three properties that most ETL work does not.</p>



<p class="wp-block-paragraph">First, results get corrected. A preliminary result goes out, the analyzer is recalibrated, and a corrected result follows hours or days later. Your pipeline is not append-only in the way a clickstream is. It has to carry amendment semantics all the way through, and an analytics table that silently keeps the first value is worse than no table at all.</p>



<p class="wp-block-paragraph">Second, the wire format is old and the transport is older. Most laboratory information systems still emit HL7 version 2 messages, pipe-delimited segments carried over the Minimum Lower Layer Protocol, which is framed bytes over a raw TCP socket. AWS documents MLLP as the common transport standard for HL7 v2 interoperability in its Healthcare Industry Lens. It has no built-in authentication and no built-in encryption. Whatever security it has, you bolt on around it.</p>



<p class="wp-block-paragraph">Third, the identifiers are load-bearing. Accession numbers, specimen IDs and medical record numbers are how the lab reconciles anything, so PHI is not a column you can drop. It is structural, and it will try to escape into every metadata surface you own.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Boundary one: getting messages out of the lab</h2>



<p class="wp-block-paragraph">The instrument or LIS sits on a network you probably do not control, run by a vendor who may not let you touch the configuration. That constrains the design more than anything on the AWS side.</p>



<p class="wp-block-paragraph">You get three realistic shapes, and the LIS vendor usually picks for you.</p>



<ul class="wp-block-list">
<li><strong>Live MLLP over a private link.</strong> An interface engine in the lab holds the TCP session and forwards over a Site-to-Site VPN or Direct Connect into a private subnet. Lowest latency, highest operational burden, because a dropped socket is a clinical incident and not a retry.</li>

<li><strong>Batch file drop over SFTP.</strong> The LIS writes flat files on a schedule and something picks them up. Far easier to reason about, far easier to audit, and acceptable whenever the downstream use is analytics rather than result delivery.</li>

<li><strong>An outbound API push.</strong> Newer systems will POST to an endpoint you expose. Best case, because you control authentication and you get a response code back, but rarest in practice.</li>
</ul>



<p class="wp-block-paragraph">Whichever shape you get, the same rule applies at this boundary: <strong>terminate the untrusted protocol as early as possible and convert to something you can authorize.</strong> A raw MLLP socket carries no identity. The moment those bytes land, wrap them in a message you can attribute, sign, and trace.</p>



<p class="wp-block-paragraph">Two things go wrong here. First, people treat a network tunnel as authentication. A VPN tells you the packets came from the lab&#8217;s network. It does not tell you which system sent them, or that the sender was authorized to send that patient&#8217;s results. Second, people assume consumer VPN products fill the gap. Tools like NordVPN or Surfshark are fine for your own admin laptop on untrusted wifi, but they carry no business associate agreement and they are not a substitute for AWS Site-to-Site VPN or AWS Client VPN on a clinical path.</p>



<p class="wp-block-paragraph">One piece of housekeeping that is cheap to do and expensive to skip: the business associate addendum. AWS makes this self-service through AWS Artifact, and it can be accepted across an entire organization. Only services on the current HIPAA Eligible Services Reference may hold PHI, and that list changes, so check it against the services you actually chose rather than your memory of it.</p>



<h2 class="wp-block-heading">Boundary two: the landing zone</h2>



<p class="wp-block-paragraph">Raw messages land in S3. This part is well documented, so I will focus on the parts people configure once and never verify.</p>



<h3 class="wp-block-heading">Make the bucket policy do the enforcing</h3>



<p class="wp-block-paragraph">Default bucket encryption means objects get encrypted. It does not mean a caller cannot write an object with different settings, and it does not stop a plaintext HTTP request. Two explicit denials cover the gap. The first protects the transport, the second protects the storage:</p>



<pre class="wp-block-code"><code>{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Sid": "DenyInsecureTransport",
      "Effect": "Deny",
      "Principal": "*",
      "Action": "s3:*",
      "Resource": [
        "arn:aws:s3:::lab-raw-inbound",
        "arn:aws:s3:::lab-raw-inbound/*"
      ],
      "Condition": {
        "Bool": { "aws:SecureTransport": "false" }
      }
    },
    {
      "Sid": "DenyWrongKey",
      "Effect": "Deny",
      "Principal": "*",
      "Action": "s3:PutObject",
      "Resource": "arn:aws:s3:::lab-raw-inbound/*",
      "Condition": {
        "StringNotEquals": {
          "s3:x-amz-server-side-encryption-aws-kms-key-id":
            "arn:aws:kms:REGION:ACCOUNT:key/KEY-ID"
        }
      }
    }
  ]
}</code></pre>



<p class="wp-block-paragraph">Use a customer managed KMS key, not the AWS managed one. The reason is not stronger cryptography, it is the key policy. A customer managed key gives you a second, independent authorization surface: even a principal with broad S3 permissions cannot read the object if the key policy does not grant them decrypt. That separation is what makes a blast radius argument credible to an assessor, and it is what makes crypto shredding possible later.</p>



<h3 class="wp-block-heading">Your object keys are metadata, and metadata leaks</h3>



<p class="wp-block-paragraph">This is the one I would put on a poster. Object contents are encrypted. Object <em>keys</em> are not. They show up in CloudTrail data events, S3 server access logs, S3 Inventory reports, bucket listings, Lambda event payloads, and every console screenshot anyone pastes into a ticket.</p>



<pre class="wp-block-code"><code># Leaks PHI into six places that are not the bucket
raw/2024/ORU_MRN4471902_SMITH_JOHN_A1C.hl7

# Same object, no identifiers in the key path
raw/dt=&lt;partition&gt;/src=lis01/msg=01HXYZ...ULID.hl7</code></pre>



<p class="wp-block-paragraph">Use an opaque, sortable identifier and keep the mapping from identifier to patient inside an encrypted store you actually control. Partition on ingest date and source system, never on anything derived from the patient. This one decision removes more PHI surface than any other single change in the pipeline.</p>



<h3 class="wp-block-heading">Immutability, and the trap inside it</h3>



<p class="wp-block-paragraph">Clinical laboratories in the United States operating under CLIA have record retention obligations set out in 42 CFR 493.1105. The floor for test reports is at least two years after the date of reporting, with pathology test reports at ten years, and several state requirements sit above the federal floor. Retention is therefore not something you invent; go and read the section, then check your state.</p>



<p class="wp-block-paragraph">S3 Object Lock gives you write-once-read-many enforcement, and it is the right tool. But understand the two modes before you commit, because one of them is genuinely irreversible.</p>



<ul class="wp-block-list">
<li><strong>Governance mode</strong> blocks deletion for most principals, but anyone holding <code>s3:BypassGovernanceRetention</code> can override it by sending the <code>x-amz-bypass-governance-retention:true</code> header. Note the S3 console includes that header by default.</li>

<li><strong>Compliance mode</strong> cannot be overridden by anyone, including the account root. Per AWS documentation, the only way to delete an object under compliance retention before its date expires is to delete the AWS account.</li>
</ul>



<p class="wp-block-paragraph">A typo in a retention date under compliance mode is permanent, and you pay to store the mistake for its full term. Object Lock also has to be enabled at bucket creation, alongside versioning. Start in governance mode, prove the retention values against real traffic, then move to compliance mode deliberately rather than as a default.</p>



<pre class="wp-block-code"><code># Indefinite hold for a specific object version, e.g. under litigation
aws s3api put-object-legal-hold 
  --bucket lab-raw-inbound 
  --key raw/dt=.../msg=01HXYZ.hl7 
  --legal-hold Status=ON</code></pre>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Boundary three: where PHI escapes into your observability</h2>



<p class="wp-block-paragraph">Here is the section that justifies the post. Everything above is on the compliance checklist. This is the part that is invisible until an audit or an incident makes it visible.</p>



<p class="wp-block-paragraph">Your operational plane is a PHI store you did not declare. Specifically:</p>



<ul class="wp-block-list">
<li><strong>Application logs.</strong> One <code>print(record)</code> left in a parser during debugging writes patient demographics into CloudWatch Logs, where the default is often no expiry at all.</li>

<li><strong>Exception traces.</strong> Unhandled parse errors frequently include the offending segment in the message. That segment is the PHI.</li>

<li><strong>Step Functions execution history.</strong> If you pass message bodies between states as payloads, that content is retained in the execution record and readable by anyone who can view executions.</li>

<li><strong>Dead letter queues.</strong> The whole point of a DLQ is to keep the failed message. A DLQ full of unparseable HL7 is a PHI archive with a different access policy than the bucket you designed so carefully.</li>

<li><strong>Third-party monitoring.</strong> Anything shipped to an external platform leaves your boundary. Grafana Cloud, Datadog and equivalents are perfectly reasonable choices for infrastructure telemetry, but a log line containing PHI going to a vendor without an executed BAA is a disclosure, not a metric.</li>

<li><strong>Notifications.</strong> An SNS topic that emails an on-call engineer &#8220;failed to process record for J. Smith, DOB &#8230;&#8221; has just sent PHI to a mail server nobody reviewed.</li>
</ul>



<h3 class="wp-block-heading">The pattern that fixes it: pass pointers, not payloads</h3>



<p class="wp-block-paragraph">Adopt a single rule and enforce it in review: <strong>PHI moves by reference between components, never by value.</strong></p>



<ol class="wp-block-list">
<li>The ingest function writes the raw message to the encrypted bucket and emits only an opaque object key plus a message ID.</li>

<li>Every downstream state, queue and event carries that pointer. Nothing carries the record body.</li>

<li>Error handling logs the pointer and a failure class, never the content that failed.</li>

<li>A DLQ therefore contains pointers to failures, and reprocessing means re-reading from the bucket under the same key policy as everything else.</li>
</ol>



<p class="wp-block-paragraph">The cost of this pattern is a slightly noisier debugging experience. You cannot read the failing record straight out of the queue. You have to go and fetch it, with credentials, and that fetch is logged. That inconvenience is the control working.</p>



<h3 class="wp-block-heading">Then close the log surface itself</h3>



<p class="wp-block-paragraph">Log groups created implicitly by Lambda and other services often retain data indefinitely unless you set a policy. Find the ones nobody configured:</p>



<pre class="wp-block-code"><code># List log groups with no retention configured
aws logs describe-log-groups 
  --query 'logGroups[?retentionInDays==null].logGroupName' 
  --output text

# Set an explicit retention on one of them
aws logs put-retention-policy 
  --log-group-name /aws/lambda/hl7-ingest 
  --retention-in-days 30</code></pre>



<p class="wp-block-paragraph">Run the first command as a scheduled check, not a one-off. New functions create new log groups, and the default comes back every time someone deploys.</p>



<p class="wp-block-paragraph">Amazon Macie is worth pointing at your buckets as a detective backstop. It will not stop a leak, but it will tell you when something started landing where it should not.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Boundary four: from identified data to analytics</h2>



<p class="wp-block-paragraph">At some point someone wants to query the data. This is the crossing where teams are most likely to convince themselves they have done something they have not.</p>



<p class="wp-block-paragraph">Amazon Comprehend Medical will detect PHI entities in unstructured clinical text through its <code>DetectPHI</code> operation, returning entity types such as name, address, identifier and date, each with a confidence score. It is a genuinely useful tool for narrative fields: pathology comments, specimen notes, microbiology free text.</p>



<pre class="wp-block-code"><code>aws comprehendmedical detect-phi 
  --text "Specimen received from the referring clinic on the stated date."</code></pre>



<p class="wp-block-paragraph">Now the part the marketing pages underplay, straight from the AWS developer guide: the entities Comprehend Medical detects <em>do not map one to one</em> to the identifier list specified by the Safe Harbor method, and AWS explicitly recommends additional human review or other methods to confirm accuracy for compliance use cases.</p>



<p class="wp-block-paragraph">Read that as: PHI detection is not de-identification. Under the HIPAA Privacy Rule, de-identification has two defined routes, Safe Harbor and Expert Determination. An NLP model with confidence scores is an input to either, not a substitute for either. If your analytics tier is meant to hold de-identified data, someone has to own that determination, and it will not be your pipeline.</p>



<p class="wp-block-paragraph">What works structurally:</p>



<ul class="wp-block-list">
<li>Put the analytics tier in a <strong>separate AWS account</strong> with its own KMS key. Not a separate bucket. A separate account, so the blast radius argument is enforced by IAM boundaries rather than by convention.</li>

<li>Make the flow <strong>one directional</strong>. The de-identification job reads from the identified side and writes to the analytics side. Nothing in the analytics account holds permissions pointing back.</li>

<li><strong>Replace identifiers with pseudonyms</strong> rather than deleting them, keeping the crosswalk in the identified account. This preserves your ability to join across results and to re-identify under a documented process, which the research use case usually needs.</li>

<li>Watch <strong>quasi-identifiers</strong>. Rare test panels, unusual reference ranges and precise timestamps re-identify people in small populations even after the obvious fields are gone. This is a statistics problem, not an IAM problem.</li>

<li>Use <strong>Lake Formation</strong> or equivalent for column and row filtering if analysts need partial access, so the restriction lives with the catalog rather than in each query tool.</li>
</ul>



<h2 class="wp-block-heading">Boundary five: sending results back out</h2>



<p class="wp-block-paragraph">Delivery is where clinical correctness and security pull against each other, and security usually loses quietly.</p>



<p class="wp-block-paragraph">The correction problem is the sharp edge. HL7 result messages carry a status that distinguishes preliminary, final and corrected results. If your delivery layer is idempotent on message ID alone, a corrected result with a new ID for the same observation will happily land beside the original instead of superseding it. Two conflicting values, both marked delivered, nobody alerted.</p>



<p class="wp-block-paragraph">Build the idempotency key from the identity of the <em>observation</em>, which typically means the placer or filler order identifier combined with the specific observation identifier, and carry the result status as a first-class field so a correction overwrites rather than appends. Keep every version, expose the current one.</p>



<p class="wp-block-paragraph">For the outbound path itself:</p>



<ul class="wp-block-list">
<li>Presigned URLs are convenient for report PDFs and they are also a bearer credential. Anyone holding the link has the document. Keep expiry short and treat generation as an auditable event.</li>

<li>If you expose a provider-facing portal, put a WAF in front of it. AWS WAF and Cloudflare are both credible here; the deciding factor is usually where the rest of your edge already lives, not the rule engines.</li>

<li>Never email a result. Email a notification that a result is available, behind authentication. The distinction sounds pedantic right up until someone forwards a mailbox.</li>
</ul>



<h2 class="wp-block-heading">Boundary six: deletion, and the conflict nobody plans for</h2>



<p class="wp-block-paragraph">Two obligations point in opposite directions. Retention rules say keep the record. Privacy rights and internal policy say be able to remove data. If you set compliance-mode Object Lock across everything, you have chosen one side without noticing.</p>



<p class="wp-block-paragraph">Resolve it by classifying before you lock. Records that fall under laboratory retention obligations go into the immutable tier with a retention period derived from the applicable rule. Derived artifacts, caches, intermediate parquet, analytics extracts and enrichment outputs do not belong there. They belong in buckets with lifecycle rules, and they are what you actually delete.</p>



<p class="wp-block-paragraph">Crypto shredding covers the middle ground: if a data set is encrypted under a key used for nothing else, scheduling that key for deletion renders the ciphertext unrecoverable without touching the objects. It is clean, it is verifiable, and it only works if you planned the key granularity up front. Retrofitting per-tenant or per-cohort keys onto a pipeline that used one key for everything is a full re-encryption exercise.</p>



<p class="wp-block-paragraph">Do not forget the endpoints. Instrument workstations, the interface engine box, and the laptop somebody used to test the parser all accumulate copies. Cloud controls do nothing for physical media, and secure erase tooling such as O&amp;O SafeErase exists for exactly that step in a decommissioning runbook.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Troubleshooting a medical lab data pipeline on AWS</h2>



<h3 class="wp-block-heading">Messages arrive but nothing appears downstream</h3>



<p class="wp-block-paragraph">Check the DLQ first, then check whether the DLQ itself has a consumer. A surprising number of pipelines have a correctly configured dead letter queue that nobody monitors, so failures accumulate silently and the only symptom is a gap in the data. Alarm on DLQ depth greater than zero, not on a threshold.</p>



<h3 class="wp-block-heading">Access denied from a Lambda that has the right IAM policy</h3>



<p class="wp-block-paragraph">With a customer managed KMS key, IAM permission on S3 is only half the grant. The function&#8217;s role also needs decrypt permission on the key, and the key policy has to allow it. If you are crossing accounts, both the key policy and the role policy must permit the action. This is the intended behavior of the two-lock design, and it is the single most common support question on any pipeline built this way.</p>



<h3 class="wp-block-heading">Deletion fails on an object you are certain you should be able to delete</h3>



<p class="wp-block-paragraph">Check for a legal hold before you check retention. A legal hold has no expiry and is independent of the retention period, so an object can be past its retention date and still undeletable. Also confirm which mode the retention uses, because governance and compliance produce the same error to a caller without bypass permission.</p>



<h3 class="wp-block-heading">Counts do not reconcile with the LIS</h3>



<p class="wp-block-paragraph">Compare on observations, not messages. One message can carry several results, corrections create additional messages for the same observation, and a naive message count will diverge from the LIS by exactly the amount that matters clinically. Reconcile daily and alert on drift rather than investigating at quarter end.</p>



<h3 class="wp-block-heading">Timestamps drift between systems</h3>



<p class="wp-block-paragraph">HL7 v2 timestamps do not always carry a timezone offset, and analyzers are often set to local time with no daylight saving handling. Capture the offset at ingest from the source system configuration and store everything in UTC with the original string preserved. Recovering an ambiguous timestamp after the fact is unpleasant, and around a daylight saving transition it may not be possible at all.</p>



<h2 class="wp-block-heading">Common mistakes</h2>



<ul class="wp-block-list">
<li><strong>Treating &#8220;HIPAA eligible&#8221; as &#8220;compliant.&#8221;</strong> Eligibility means AWS will cover the service under the BAA. Configuration is still entirely yours.</li>

<li><strong>Putting identifiers in object keys, queue names, or log group names.</strong> Encrypted contents, plaintext metadata.</li>

<li><strong>Enabling compliance-mode Object Lock before validating retention values.</strong> There is no undo, and no support ticket that fixes it.</li>

<li><strong>Assuming automated PHI detection equals de-identification.</strong> AWS itself says the entity list does not map one to one to Safe Harbor identifiers.</li>

<li><strong>Running the analytics tier in the same account as identified data.</strong> A bucket boundary is a policy away from collapsing. An account boundary is not.</li>

<li><strong>Shipping application logs to an external platform without checking what the log lines contain.</strong> Test it by grepping your own log group for a known test patient name.</li>

<li><strong>Ignoring corrected results until an analyst notices.</strong> Amendment handling is a day-one requirement, not a phase two feature.</li>
</ul>



<h2 class="wp-block-heading">Best practices worth the effort</h2>



<ul class="wp-block-list">
<li><strong>Write the data flow diagram first and mark every trust boundary crossing.</strong> Every arrow that crosses one needs a named control. This exercise finds more problems than any scanner.</li>

<li><strong>Encode the eligible service list as policy.</strong> Service control policies at the organization level stop someone from putting PHI into a service you never assessed.</li>

<li><strong>Keep VPC endpoints on the PHI path.</strong> Interface and gateway endpoints keep traffic to AWS services off the public internet and give you an endpoint policy as an extra choke point. Check which of your chosen services offer one.</li>

<li><strong>Build with synthetic HL7 from day one.</strong> Nobody should need production PHI to develop a parser. A cheap VPS from Contabo or InterServer is fine for a synthetic-data development box, and keeping that environment entirely outside the PHI boundary is the point.</li>

<li><strong>Alarm on absence.</strong> A feed that stops is more dangerous than a feed that errors, because errors are loud. Alert when expected message volume for a source drops below its floor.</li>

<li><strong>Test restores, not backups.</strong> Restore into an isolated account, confirm the KMS grants work there, and document how long it took.</li>

<li><strong>Review CloudTrail for data events on the PHI buckets.</strong> Object-level logging costs money and is the only record of who read what.</li>
</ul>



<h2 class="wp-block-heading">FAQ</h2>



<h3 class="wp-block-heading">Do I need a BAA with AWS before I start building?</h3>



<p class="wp-block-paragraph">Before PHI touches the environment, yes. You can architect and test with synthetic data first. AWS provides the business associate addendum through AWS Artifact as a self-service acceptance that can be applied across an organization, and only services listed on the HIPAA Eligible Services Reference may process, store or transmit PHI.</p>



<h3 class="wp-block-heading">Should I use AWS HealthLake or build on S3, Glue and Athena?</h3>



<p class="wp-block-paragraph">HealthLake is a managed FHIR datastore with FHIR APIs, medical NLP and query built in, and it is the shorter path if your consumers speak FHIR. It does not natively ingest HL7 v2; AWS points to partner tooling or your own transformation for non-FHIR input. If your consumers are analysts with SQL and the data is structured lab results rather than clinical documents, an S3 and Athena lakehouse is simpler and easier to keep inside a boundary you control. Decide on the consumer, not the source.</p>



<h3 class="wp-block-heading">How long do I have to keep laboratory records?</h3>



<p class="wp-block-paragraph">For CLIA-regulated laboratories in the United States, 42 CFR 493.1105 sets the federal floor, including at least two years for test reports after the date of reporting and at least ten years for pathology test reports. State law frequently exceeds these minimums, so the applicable period is whichever is longer for your jurisdiction. Read the regulation and confirm with your compliance lead before you encode a number into a retention policy.</p>



<h3 class="wp-block-heading">Can Amazon Comprehend Medical de-identify data for me?</h3>



<p class="wp-block-paragraph">It can detect PHI entities and give you confidence scores, which is a strong starting point for redaction workflows. It does not by itself satisfy HIPAA&#8217;s de-identification standard. AWS documents that the detected entities do not map one to one to the Safe Harbor identifier list and recommends human review for compliance use cases. Treat it as a detector feeding a documented Safe Harbor or Expert Determination process.</p>



<h3 class="wp-block-heading">Is S3 Object Lock compliance mode required for lab records?</h3>



<p class="wp-block-paragraph">Not inherently. The argument for compliance mode is a threat model in which a compromised or malicious administrator could remove retention, not a regulation that names the mode. Governance mode plus tightly controlled bypass permissions and strong audit logging is a defensible position for many laboratories. Choose deliberately and write down the reasoning, because compliance mode commits you to the storage cost for the full term with no exit.</p>



<h3 class="wp-block-heading">Where does PHI most often leak in a pipeline that looks correctly configured?</h3>



<p class="wp-block-paragraph">Logs, dead letter queues, workflow execution histories, notification messages and object key names. All five are metadata surfaces that sit outside the storage layer everyone reviews. Grep your own log groups for a known test patient identifier and you will find out quickly whether your pipeline has the problem.</p>



<h3 class="wp-block-heading">Do I need a dedicated AWS account just for the lab pipeline?</h3>



<p class="wp-block-paragraph">Separate accounts for identified and de-identified data are worth it even at small scale, because the boundary is then enforced by IAM rather than by naming conventions. Whether the PHI workload also needs isolating from your other production workloads depends on who holds admin access there. If the answer is &#8220;broadly the same people,&#8221; separate it.</p>



<h2 class="wp-block-heading">The one thing to take away</h2>



<p class="wp-block-paragraph">A secure medical lab data pipeline on AWS is not defined by the encryption on its storage layer. Encryption is the easy part, and it is the part everyone gets right. What separates a design that survives an audit from one that does not is whether PHI can travel by value through your operational plane: into logs, into queue bodies, into workflow histories, into filenames, into alert emails.</p>



<p class="wp-block-paragraph">Move PHI by reference, keep identifiers out of every metadata surface, and put an account boundary between identified and de-identified data. Do those three things and the rest of the compliance work becomes documentation rather than redesign.</p>



<h2 class="wp-block-heading">Work with me on healthcare data pipelines</h2>



<p class="wp-block-paragraph">I design and review AWS data platforms for regulated workloads, and lab data is a specific enough problem that generic cloud advice tends to miss it. Things I can help with:</p>



<ul class="wp-block-list">
<li>Mapping every trust boundary in an existing pipeline and finding the PHI surfaces nobody documented, including logs, queues and object key schemes.</li>

<li>Designing HL7 v2 or file-drop ingestion into S3 with KMS key separation, VPC endpoints and bucket policies that enforce rather than suggest.</li>

<li>Building the identified and de-identified account split, including the one-way transformation job and the pseudonym crosswalk.</li>

<li>Getting amendment and correction handling right so corrected results supersede rather than duplicate, all the way to the analytics tables.</li>

<li>Retention and deletion design: Object Lock modes, lifecycle policies, crypto shredding key granularity, and the runbook that ties them together.</li>

<li>Observability that is useful without being a disclosure risk, including CloudWatch retention hygiene and safe alerting patterns.</li>
</ul>



<p class="wp-block-paragraph">If you have an architecture diagram, a Terraform plan, or a redacted sample of the messages you are receiving, send it over and I will tell you what I would change first.</p>



<div class="wp-block-buttons is-layout-flex wp-block-buttons-is-layout-flex">
<div class="wp-block-button"><a class="wp-block-button__link wp-element-button" href="https://www.upwork.com/freelancers/~01f15a912ad84a6620" target="_blank" rel="noreferrer noopener">Work with me on Upwork</a></div>
</div>
<p>The post <a href="https://john-nessime.com/blog/cloud-computing/medical-lab-data-pipeline-aws/">Where PHI Escapes: Building a Secure Medical Lab Data Pipeline on AWS</a> appeared first on <a href="https://john-nessime.com/blog">John Nessime</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>Build a Secure AI Medical Assistant on AWS: The Boundaries That Actually Leak</title>
		<link>https://john-nessime.com/blog/case-studies/secure-ai-medical-assistant-aws/</link>
		
		<dc:creator><![CDATA[John Nessime]]></dc:creator>
		<pubDate>Mon, 03 Aug 2026 13:45:12 +0000</pubDate>
				<category><![CDATA[Case Studies]]></category>
		<category><![CDATA[Cloud Computing]]></category>
		<category><![CDATA[DevOps]]></category>
		<category><![CDATA[Technical Guides]]></category>
		<category><![CDATA[Amazon Bedrock]]></category>
		<category><![CDATA[Amazon Comprehend Medical]]></category>
		<category><![CDATA[Amazon S3]]></category>
		<category><![CDATA[Architecture]]></category>
		<category><![CDATA[AWS]]></category>
		<category><![CDATA[Bedrock Guardrails]]></category>
		<category><![CDATA[Cloud Security]]></category>
		<category><![CDATA[CloudWatch]]></category>
		<category><![CDATA[Data Residency]]></category>
		<category><![CDATA[Generative AI]]></category>
		<category><![CDATA[Healthcare AI]]></category>
		<category><![CDATA[HIPAA]]></category>
		<category><![CDATA[IAM]]></category>
		<category><![CDATA[Logging]]></category>
		<category><![CDATA[PHI]]></category>
		<category><![CDATA[RAG]]></category>
		<category><![CDATA[Vector Database]]></category>
		<category><![CDATA[VPC]]></category>
		<guid isPermaLink="false">https://john-nessime.com/blog/?p=127</guid>

					<description><![CDATA[<p>A practical architecture for a secure AI medical assistant on AWS, organised by the boundary the data crosses: the input box, your own invocation logs, cross-Region inference routing, the retrieval index, and clinical accuracy. Includes real commands, the failure modes that stay invisible until an audit, and the trade-offs worth knowing before you build.</p>
<p>The post <a href="https://john-nessime.com/blog/case-studies/secure-ai-medical-assistant-aws/">Build a Secure AI Medical Assistant on AWS: The Boundaries That Actually Leak</a> appeared first on <a href="https://john-nessime.com/blog">John Nessime</a>.</p>
]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">The message usually arrives on a Friday afternoon: &#8220;One of the residents pasted a real discharge summary into the demo.&#8221; Nobody meant anything by it. The thing was a study aid, a chat box over a pile of reference material, running in a sandbox account with no Business Associate Addendum in place and invocation logging switched on because logging is a good habit. And now there is protected health information sitting in plaintext in a CloudWatch log group in an account that was never in scope for it.</p>



<p class="wp-block-paragraph">That is the failure mode worth planning for. Not a jailbreak, not a model saying something clinically wrong on stage. A user typing something perfectly reasonable into a box you built, and the data ending up somewhere you never drew on the diagram. This post walks through how to build a secure AI medical assistant on AWS, organised by the boundary the data actually crosses: the input box, your own logs, the Region the inference runs in, the retrieval index, and finally the answer itself. Commands are included where they explain something. Where a value depends on your account or your counsel, I say so instead of making one up.</p>



<h2 class="wp-block-heading">Two different products hiding behind one request</h2>



<p class="wp-block-paragraph">&#8220;An AI assistant for medical students and doctors&#8221; is two builds with two risk profiles, and conflating them is the root of most of the trouble.</p>



<ul class="wp-block-list">
<li><strong>The study tool.</strong> Question banks, guideline summaries, differential drills, spaced repetition. In theory it never touches patient data. Its real risk is confident wrongness and unattributed answers, not privacy.</li>

<li><strong>The clinical assistant.</strong> Note summarisation, coding support, chart question answering. It handles PHI by design, so the whole thing has to sit inside a HIPAA-designated account from day one.</li>
</ul>



<p class="wp-block-paragraph">The trap is that the study tool becomes the clinical assistant without anyone shipping a release. A student rehearses a case they saw on the ward. A doctor tries the study tool on a real chart because it is the one that is already open. The moment your input box accepts free text from someone with clinical access, you should assume PHI will arrive in it. Build accordingly, or put the study tool on infrastructure where PHI arriving is survivable.</p>



<p class="wp-block-paragraph">My default is to run both in the same HIPAA-designated account with the same controls, and keep only the marketing site, the docs and the waitlist form somewhere ordinary and cheap like InterServer or any commodity host, entirely outside the AWS organisation. Small blast radius beats clever separation you have to explain to an auditor.</p>



<h2 class="wp-block-heading">Boundary one: PHI arrives before you decide to accept it</h2>



<p class="wp-block-paragraph">Start with the thing that trips up almost every first build: HIPAA-eligible and HIPAA-compliant are not the same word. AWS designating a service as HIPAA-eligible means you are permitted to process PHI with it under an executed Business Associate Addendum. It says nothing about whether your deployment is compliant. That part is entirely yours.</p>



<p class="wp-block-paragraph">Three conditions have to hold together before any PHI touches a service: the service is on the AWS HIPAA Eligible Services Reference, you have an executed BAA, and the account is designated for HIPAA use. Amazon Bedrock and Amazon Comprehend Medical both appear on that list, and the BAA is accepted through AWS Artifact rather than negotiated by email. Check the reference page yourself before you commit to an architecture, because the list changes and every component in your diagram has to be on it, not just the model.</p>



<h3 class="wp-block-heading">Detecting PHI at the door</h3>



<p class="wp-block-paragraph">Amazon Comprehend Medical has an operation specifically for finding protected health information in unstructured clinical text. You hand it free text, it returns detected entities with a type, a confidence score and character offsets.</p>



<pre class="wp-block-code"><code>aws comprehendmedical detect-phi 
  --region us-east-1 
  --text "Patient seen for chest pain, MRN 004512, discharged Tuesday."</code></pre>



<p class="wp-block-paragraph">The offsets are the useful part. They let you redact or tokenise in place before the text goes anywhere else, rather than throwing the whole message away and telling the user to try again.</p>



<p class="wp-block-paragraph">Now the caveat that matters more than the feature. AWS states plainly that Comprehend Medical may not identify PHI in all circumstances and that it does not, on its own, meet HIPAA&#8217;s requirements for de-identification. Read that as: it is a good filter and a terrible guarantee. If your compliance story is &#8220;we strip PHI before the model sees it, so we are outside HIPAA scope,&#8221; that story does not hold. Treat detection as defence in depth inside a compliant account, not as an escape hatch from needing one.</p>



<h2 class="wp-block-heading">Boundary two: your own logs are the most likely leak</h2>



<p class="wp-block-paragraph">This is the invisible one. Bedrock model invocation logging is disabled by default and captures the full request data, response data and metadata for every call in the account, in that Region. You turn it on for a good reason, usually because someone in security asked who prompted what and when. Then it quietly becomes the largest concentration of raw clinical text you own, and it is nowhere on the architecture diagram because it is a checkbox rather than a component.</p>



<p class="wp-block-paragraph">Check whether it is on before you assume anything:</p>



<pre class="wp-block-code"><code>aws bedrock get-model-invocation-logging-configuration --region us-east-1</code></pre>



<p class="wp-block-paragraph">The configuration is per account per Region, so run it in every Region where anyone has ever opened the Bedrock console. Setting it deliberately looks like this:</p>



<pre class="wp-block-code"><code>aws bedrock put-model-invocation-logging-configuration 
  --region us-east-1 
  --logging-config '{
    "s3Config": {
      "bucketName": "med-assistant-invocation-logs",
      "keyPrefix": "bedrock/"
    },
    "textDataDeliveryEnabled": true,
    "imageDataDeliveryEnabled": false,
    "embeddingDataDeliveryEnabled": false
  }'</code></pre>



<p class="wp-block-paragraph">Each delivery flag is a separate decision about a separate category of PHI. Text is the obvious one. Image delivery matters the moment anyone uploads a photographed chart or a scan, because burned-in identifiers travel with the pixels and no text filter will ever see them. Embedding delivery is the one people leave on without thinking; vectors derived from clinical text are not a safe artefact, and they are large.</p>



<p class="wp-block-paragraph">The sharpest detail is buried in the Guardrails documentation: AWS notes you can disable invocation logs if you do not want blocked content appearing as plaintext in them. Read the implication. A guardrail can refuse a prompt, mask the identifiers, and stop the model ever seeing them, and the original text still lands in your log destination. The guardrail protects the model call. It does not protect the log.</p>



<ul class="wp-block-list">
<li>Encrypt the log destination with a customer-managed KMS key, and keep the key policy tight enough that &#8220;everyone with S3 read&#8221; is not also &#8220;everyone with chart access&#8221;.</li>

<li>Set a retention period that reflects a legal decision someone actually made, not the CloudWatch default of never expiring.</li>

<li>Send logs to a separate, tightly scoped account if your organisation is large enough to have people who need dashboards but not records.</li>

<li>Remember your application logs too. A framework that logs request bodies on error will do this to you long before Bedrock does.</li>
</ul>



<h2 class="wp-block-heading">Boundary three: where the inference actually runs</h2>



<p class="wp-block-paragraph">Cross-Region inference in Bedrock exists because on-demand capacity is uneven and bursts happen. It routes your request to another Region to get it served. There are two flavours and the difference is not cosmetic.</p>



<ul class="wp-block-list">
<li><strong>Geographic profiles</strong> keep routing inside a defined geography such as the US or the EU. A request that starts in the EU stays in EU Regions. This is the one built for residency requirements.</li>

<li><strong>Global profiles</strong> route to supported commercial Regions worldwide for maximum throughput. AWS documents that a request can be routed to a destination Region even if you never opted that Region into your account.</li>
</ul>



<p class="wp-block-paragraph">To be fair to global profiles: data is not stored in the destination Region, transfer happens encrypted across the AWS network, and your invocation logs, knowledge bases and configuration all stay in the source Region. For a workload with no geographic constraint it is a genuinely good default, and it typically carries a lower per-token rate than staying in-geography. For a clinical workload with a residency commitment in a contract, it is the wrong tool, and &#8220;the prompt left the geography but was not stored there&#8221; is a sentence you do not want to be constructing during an audit.</p>



<p class="wp-block-paragraph">Pin it in policy rather than trusting a config value in a repo. A Service Control Policy denying the Bedrock API outside your approved Regions closes both doors at once, because invoking a cross-Region profile requires model access in the destination Regions as well as the source:</p>



<pre class="wp-block-code"><code>{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Sid": "DenyBedrockOutsideApprovedRegions",
      "Effect": "Deny",
      "Action": "bedrock:*",
      "Resource": "*",
      "Condition": {
        "StringNotEquals": {
          "aws:RequestedRegion": ["us-east-1", "us-west-2"]
        }
      }
    }
  ]
}</code></pre>



<p class="wp-block-paragraph">The trade-off is real and you should know it before you apply this. If you later adopt a geographic profile whose destination list includes a Region you denied, invocations fail with an access error that looks nothing like a routing problem. Keep the approved Region list and the profile&#8217;s destination list in the same review, not in two different tickets.</p>



<h2 class="wp-block-heading">Boundary four: retrieval, and the grounding check that will not save you</h2>



<p class="wp-block-paragraph">A useful medical assistant is retrieval-augmented. The model alone is a fluent generalist; the value comes from grounding answers in a curated corpus, whether that is institutional guidelines, formulary rules or a licensed reference set. Bedrock Knowledge Bases will manage the ingestion and the vector store for you, or you can run your own index in Amazon OpenSearch Service.</p>



<p class="wp-block-paragraph">Two things bite here.</p>



<p class="wp-block-paragraph"><strong>Tenant isolation in the index.</strong> If you serve more than one hospital, department or study cohort, a shared index with a filter applied in application code is one refactor away from cross-tenant retrieval. Filters are easy to forget and impossible to notice, because a wrong answer that cites a real document looks exactly like a right one. Separate indexes per tenant, or fine-grained access control enforced below your application, cost more and fail safe.</p>



<p class="wp-block-paragraph"><strong>The grounding check has a scope limit.</strong> Guardrails contextual grounding checks detect responses that are not supported by the source material or not relevant to the question. Excellent feature. The documentation states the supported use cases are summarisation, paraphrasing and question answering, and that conversational chatbot use cases are not supported. If your product is a chat interface with turn history, do not assume this check is covering you. There is also a streaming wrinkle: relevance is assessed per chunk, so an irrelevant response can reach the user before it is marked irrelevant at the end of the stream.</p>



<p class="wp-block-paragraph">What you can rely on regardless of interface is applying a guardrail to arbitrary text directly, which is how you check an input before it enters your own pipeline:</p>



<pre class="wp-block-code"><code>import boto3

client = boto3.client("bedrock-runtime", region_name="us-east-1")

response = client.apply_guardrail(
    guardrailIdentifier=GUARDRAIL_ID,
    guardrailVersion="DRAFT",
    source="INPUT",
    content=[{"text": {"text": user_message}}],
)

if response["action"] == "GUARDRAIL_INTERVENED":
    # Stop here. Do not forward, and do not write the raw text anywhere.
    handle_blocked(response["assessments"])</code></pre>



<p class="wp-block-paragraph">Note what the comment is doing. The most common bug in this pattern is catching the intervention and then logging the offending input &#8220;for debugging&#8221;, which reintroduces exactly the leak the guardrail just prevented. Log the assessment, log a request identifier, never the text.</p>



<p class="wp-block-paragraph">Sensitive information filters give you two handling modes: block, which rejects the request outright, and mask, which replaces detected entities with placeholder tags. For a study tool, mask is usually right, because a student who typed a name by reflex gets a useful answer instead of a wall. For a clinical assistant working over charts, blocking on unexpected identifiers in an input that should have arrived structured is a better signal that something upstream is wrong.</p>



<h2 class="wp-block-heading">Boundary five: being confidently wrong</h2>



<p class="wp-block-paragraph">Everything above is about data leaving. This one is about a bad answer arriving, and for a medical audience it is the reputational failure that actually ends products.</p>



<ul class="wp-block-list">
<li><strong>Cite or refuse.</strong> Return the retrieved passages alongside the answer. If retrieval returned nothing above your relevance threshold, say so rather than letting the model answer from parametric memory. A student cannot verify what they cannot see.</li>

<li><strong>Version the corpus, not just the model.</strong> When a guideline changes, you need to know which answers were generated against the old text. Store a corpus revision identifier with every logged response.</li>

<li><strong>Use denied topics deliberately.</strong> Dosing for a named patient, and anything that reads as a treatment directive rather than reference information, are reasonable things to route to a refusal with a clear explanation.</li>

<li><strong>Get regulatory advice early.</strong> Whether clinical decision support software is regulated as a medical device depends on your jurisdiction and, critically, on the claims you make about it. This is a legal question with engineering consequences, and it is much cheaper to answer before the interface exists.</li>
</ul>



<h2 class="wp-block-heading">A build order for a secure AI medical assistant on AWS</h2>



<p class="wp-block-paragraph">Sequence matters here more than in most builds, because several of these are painful to retrofit.</p>



<ol class="wp-block-list">
<li>Accept the BAA through AWS Artifact and designate the account for HIPAA use. Do this before the first prototype, not before the first customer.</li>

<li>Pin Regions with a Service Control Policy, and decide the geographic-versus-global inference profile question in writing.</li>

<li>Configure invocation logging deliberately, with a customer-managed KMS key, an explicit retention period, and each data-type delivery flag chosen rather than defaulted.</li>

<li>Put the application in private subnets and reach Bedrock over VPC endpoints so PHI-bearing traffic does not traverse the public internet.</li>

<li>Build the guardrail before the prompt. Sensitive information filters, denied topics, and grounding checks where they apply.</li>

<li>Add Comprehend Medical detection in the ingestion path for anything you are storing, and at the input boundary for anything a user types.</li>

<li>Build retrieval with tenant isolation from the first index, not the second.</li>

<li>Only now write the assistant&#8217;s prompt and interface, and put a WAF such as Cloudflare or AWS WAF in front of the public endpoint.</li>
</ol>



<h2 class="wp-block-heading">Troubleshooting</h2>



<ul class="wp-block-list">
<li><strong>AccessDenied on a model that clearly works elsewhere.</strong> Usually one of three things: model access not requested in this Region, a cross-Region profile whose destination Regions your SCP denies, or an IAM policy that grants the model in the source Region only.</li>

<li><strong>No invocation logs appearing.</strong> The configuration is per Region and disabled by default. Confirm you queried the same Region the application calls, and that the delivery flag for the data type you expect is enabled.</li>

<li><strong>Guardrail passes text you expected it to catch.</strong> Sensitive information detection is probabilistic and context-dependent. Very short inputs give it little to work with. Test with realistic clinical phrasing, not single tokens, and add regex patterns for structured identifiers like MRNs that follow a house format.</li>

<li><strong>Grounding check appears to do nothing.</strong> Check your interface shape against the supported use cases before assuming it is misconfigured, and check whether streaming is masking the result until the response completes.</li>

<li><strong>Retrieval returns plausible but wrong documents.</strong> Look at chunking before you look at the model. Clinical guidelines chunked mid-table or mid-criteria retrieve badly no matter what embedding you use.</li>
</ul>



<h2 class="wp-block-heading">Common mistakes</h2>



<ul class="wp-block-list">
<li>Prototyping in a personal or sandbox account and promising to migrate later. The prototype is where the first real chart gets pasted.</li>

<li>Assuming the eligibility of Bedrock covers the whole stack. Every component that touches PHI needs to be eligible, including the vector store, the queue and the cache.</li>

<li>Treating PHI detection as de-identification. AWS says explicitly that it is not.</li>

<li>Leaving image and embedding log delivery enabled by copy-paste.</li>

<li>Logging blocked prompts to debug the guardrail.</li>

<li>Shipping a chat interface and citing contextual grounding as the hallucination control.</li>

<li>Filtering tenants in application code over a shared index.</li>
</ul>



<h2 class="wp-block-heading">Best practices</h2>



<ul class="wp-block-list">
<li>Write the data-flow diagram with logs, backups and the vector index drawn as first-class destinations. If PHI can land there, it is on the diagram.</li>

<li>Define everything in Terraform or OpenTofu so the guardrail, the logging configuration and the SCP are reviewable artefacts rather than console state.</li>

<li>Keep a small evaluation set of realistic clinical questions with known-good answers, and run it on every prompt or model change.</li>

<li>Alarm on guardrail intervention rate. A sudden rise usually means a change upstream, not a change in users.</li>

<li>Dashboard invocation counts, latency and intervention rates somewhere your on-call actually looks, whether that is CloudWatch, Grafana or Datadog.</li>

<li>Scope IAM to specific model ARNs and specific guardrail identifiers. A wildcard on <code>bedrock:InvokeModel</code> means any model, including ones you never evaluated.</li>

<li>Rehearse the breach path once. Knowing which bucket, which log group and which key you would need to reason about is worth an afternoon.</li>
</ul>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Frequently asked questions</h2>



<h3 class="wp-block-heading">Is Amazon Bedrock HIPAA compliant?</h3>



<p class="wp-block-paragraph">Bedrock is HIPAA-eligible, which is a different claim. Eligibility means AWS permits you to process PHI with the service under an executed BAA. Compliance is a property of your deployment: your account designation, encryption, access control, network isolation, logging and retention. A Bedrock workload with a signed BAA and public endpoints and unbounded plaintext logs is not compliant.</p>



<h3 class="wp-block-heading">Do I need a BAA if the tool is only for medical students?</h3>



<p class="wp-block-paragraph">If the tool genuinely never receives PHI, HIPAA is not engaged. The practical question is whether you can guarantee that, given that your users have clinical access and a free-text box. If you cannot, get the BAA. It costs you a self-service acceptance in AWS Artifact and removes an entire category of incident.</p>



<h3 class="wp-block-heading">Does AWS use my prompts to train the models?</h3>



<p class="wp-block-paragraph">AWS states that customer content submitted to Bedrock is not used to train the underlying foundation models or shared with model providers. That statement is the sort of thing a hospital security review will want quoted verbatim from the current AWS data protection documentation rather than from a blog, so pull the live wording when you write your assessment.</p>



<h3 class="wp-block-heading">Is Comprehend Medical enough to de-identify clinical text?</h3>



<p class="wp-block-paragraph">No. AWS documents that it may not identify PHI in all circumstances and does not by itself meet HIPAA&#8217;s de-identification requirements. Use it as a detection layer and a redaction aid inside a compliant environment. Formal de-identification, whether by the Safe Harbor method or expert determination, is a separate exercise with its own review.</p>



<h3 class="wp-block-heading">Where should the vector store live?</h3>



<p class="wp-block-paragraph">In the same account and Region as the rest of the workload, on a HIPAA-eligible service, encrypted with a customer-managed key, reachable only from private subnets. Embeddings derived from clinical text are not sanitised data and should not be treated as a lower-sensitivity artefact than the source.</p>



<h3 class="wp-block-heading">Should I use a global or geographic inference profile?</h3>



<p class="wp-block-paragraph">Geographic if you have any residency commitment, contractual or regulatory. Global if you have none and want the throughput and the lower per-token rate. Decide once, document the reasoning, and enforce it with a Service Control Policy rather than a configuration constant.</p>



<h3 class="wp-block-heading">Does an AI medical assistant count as a medical device?</h3>



<p class="wp-block-paragraph">It depends on your jurisdiction and on what you claim the software does. Software that surfaces reference information a clinician independently reviews has generally been treated differently from software that directs a clinical decision, but the boundary is fact-specific and moves. This is a question for regulatory counsel before launch, not a question for your architecture diagram.</p>



<h2 class="wp-block-heading">The one thing worth remembering</h2>



<p class="wp-block-paragraph">A secure AI medical assistant on AWS is not mainly a model problem. Bedrock, Guardrails and Comprehend Medical are the easy part, and the documentation for them is good. The hard part is that PHI leaves through the paths you did not draw: an invocation log you enabled for good reasons, a global inference profile that routes wherever capacity exists, an index shared between tenants, a debug line added at two in the morning.</p>



<p class="wp-block-paragraph">So build the boundary first and the assistant second. Get the BAA accepted, pin the Regions in policy, decide consciously what your logs are allowed to hold, and isolate retrieval per tenant before there is a second tenant. Every one of those is cheap on day one and expensive in month six.</p>



<h2 class="wp-block-heading">Need a second pair of eyes on your build?</h2>



<p class="wp-block-paragraph">I work with teams building AI on AWS where the data is sensitive and the failure modes are quiet. Things I can help with on a project like this:</p>



<ul class="wp-block-list">
<li>Reviewing a Bedrock architecture against the boundaries above and telling you where PHI can actually land</li>

<li>Setting up account separation, BAA scope, Region pinning with Service Control Policies, and VPC endpoint access to Bedrock</li>

<li>Designing and tuning Guardrails policies, including custom regex for house identifier formats, and the block-versus-mask decision per surface</li>

<li>Building the retrieval layer with per-tenant isolation, sensible clinical chunking, and citation-or-refuse behaviour</li>

<li>Auditing invocation logging, KMS key policies, retention and application-level log hygiene for accidental PHI capture</li>

<li>Putting the whole thing in Terraform or OpenTofu so your controls are reviewable instead of remembered</li>
</ul>



<p class="wp-block-paragraph">If any of that is on your plate, send me the piece you are least sure about. A redacted architecture diagram, a guardrail configuration, a logging policy, an <code>AccessDenied</code> you cannot explain. I would rather look at the real thing than talk in generalities.</p>



<div class="wp-block-buttons is-layout-flex wp-block-buttons-is-layout-flex">
<div class="wp-block-button"><a class="wp-block-button__link wp-element-button" href="https://www.upwork.com/freelancers/~01f15a912ad84a6620" target="_blank" rel="noreferrer noopener">Work with me on Upwork</a></div>
</div>
<p>The post <a href="https://john-nessime.com/blog/case-studies/secure-ai-medical-assistant-aws/">Build a Secure AI Medical Assistant on AWS: The Boundaries That Actually Leak</a> appeared first on <a href="https://john-nessime.com/blog">John Nessime</a>.</p>
]]></content:encoded>
					
		
		
			</item>
	</channel>
</rss>
