Patients' inability to understand pathology and radiology reports is globally recognised.
When faced with technical reports they cannot decode, many patients turn to Google or ChatGPT — neither of which reliably provides clinically sound, context-aware explanations.
Stanford Medicine ran a pilot that drafts plain-language explanations of pathology results for clinicians to review and share. https://med.stanford.edu/news/all-news/2025/01/ai-test-results.html
"When a physician orders a test for a patient — be it blood work, an X-ray, a biopsy, or something else — the results appear as medical data without interpretation. By law, those results must be shared with the patient as soon as they're ready, but the descriptions are often highly technical and can be challenging to understand without a medical degree. That means physicians are tasked with translating the results for their patients. …"
"The technology drafts an interpretation of clinical test and lab results and explains them in a message using plain language, which a physician then reviews and approves…."
Harvard Medical School piloted video explanations of radiology reports that combine plain-language narration with on-image annotations. https://arxiv.org/html/2410.00441v1
"These documents often remain inaccessible to the very individuals they concern most: the patients themselves. The complex medical terminology and abbreviations, designed for efficient communication among healthcare professionals, create a significant barrier to patient understanding. This can lead to misunderstandings, anxiety, and potentially impact treatment decisions, ultimately increasing the burden on doctors. …"
The outcome is a video that: "....(1) explains the important findings in plain language; (2) points to abnormalities on the image; and (3) shows how the image would appear if it came from a healthy individual..."
What is healthQ's Pathology Explainer?
healthQ's Explainers make test results clinically meaningful by adding patient context and involving the pathologists who author the reports.
It is an AI agent that answers patients' immediate questions about their pathology report at the point they receive it from the lab.
Why are we building this?
Between a pathologist writing a report for a treating clinician and the patient receiving the report, patients are left confused about their health concerns. That communication gap is the problem we are addressing.
Findings from healthQ's own research
Our research with 350+ patients across diagnostic labs and OPD waiting areas revealed these key findings:

~87% of patients said they used ChatGPT or Google to interpret their lab results.
1. Challenges patients face:
  • Language of the report: the single biggest barrier.
  • Clinical jargon used in the report.
  • Test values vary by age and gender.
  • Not all deviations are equivalent; some are more extreme.
  • Deviations on opposite sides of the reference range do not mean the same thing.
  • Some abnormal values are incidental, while others indicate systemic disease.
  • Some values change slowly over time.
  • Some tests (e.g., NS1 or FSH/LH) depend on the timing of the sample.
  • Some tests are derived; their highlighted values can be misleading.
  • Reports rarely state the indication for testing — why the test was ordered.
  • Different values within the same panel (for example, a lipid panel) can mean very different things.
  • Patients also want to ask conceptual questions (e.g., "How does thyroid function work in my body?").
2. Treating clinicians' frustrations:
  • Managing many different report formats and extracting the relevant values.
  • Pulling together historical data from older reports.
healthQ's explainer, therefore, has the following features
1
Zero-shot report generation
  • Produces an explainer in the patient's preferred language, contextualised to the patient history.
  • Generates a non-standard, dynamic report structure that adapts to the tests present.
2
Filters
  • Change language: Hindi / English / regional languages.
  • Drop numbers: generate a report without numeric values.
  • Remove jargon: minimise clinical terminology.
  • Simpler to Advanced: adjust technical complexity (levels 1 / 2 / 3).
3
Modify a section
  • Re-explain any panel or paragraph.
  • Select any text in the report and request a re-explanation.
Agentic capabilities
1
Ask questions
  • Explain tests related to a symptom (e.g., joint pain).
  • Explain tests related to a known problem (e.g., fertility, dengue).
  • Explain a panel in detail (e.g., "How are my lipid results?").
  • Explain phenomena (e.g., "How does our body generate cholesterol?").
  • Ask any custom question by typing it.
2
Correlate with my previous report
Re-generate the explainer by incorporating values from prior reports to show trends and meaningful changes.
Made with