AI-assisted interpretation of laboratory results and pathology slides

01 / Research focus

Research questions

• Can large-scale analysis of laboratory sample results and digitised pathology slides help prioritise cases for pathologist review?
• Can interpretable flags identify cases where additional markers or parameters merit consideration?
• How can machine-learning models analyse relationships between biomarkers while making their outputs understandable to pathologists?
• How can these methods be evaluated for clinical relevance, missed findings, and unnecessary follow-up suggestions?

02 / Institutional context

NIMT Institute of Medical and Paramedical Sciences, Greater Noida

41, Knowledge Park I, Greater Noida, Uttar Pradesh 201310.

The NIMT–Seralis Diagnostics Research Lab is run under NIMT Hospital and Research Center, a nonprofit research organisation, and is the home of this research programme. Its institutional affiliation is NIMT Institute of Medical and Paramedical Sciences.

The institute appears in Atal Bihari Vajpayee Medical University, Uttar Pradesh’s official affiliated-college register. The register lists B.Sc. Nursing under college code 2010063, and Bachelor of Physiotherapy (BPT) and Bachelor in Medical Radiologic & Imaging Technology (BMRIT) under allied-health college code 2010618.

The Indian Nursing Council’s 2022–23 B.Sc. Nursing list records the institute as suitable for an annual intake of 40. The Pharmacy Council of India’s published decision for PCI-1458 extends D.Pharm approval through 2025–26 for 60 admissions, subject to inspection and faculty-registration conditions.

These dated records relate to the named educational programmes. They do not establish approval or clinical validation of the diagnostics research programme.

03 / Methodology

Research approach

The research approach involves defining clinically meaningful review criteria with pathologists, followed by preparing appropriately governed datasets of laboratory sample results and digitised pathology slides, developing computational methods, and evaluating their outputs against expert assessment. The intended scale requires reproducible processing across large numbers of samples and slide images; dataset size and availability remain subject to data-access permissions and required approvals.

The programme focuses on computational analysis of laboratory result data and digitised images of pathology slides. Development of new blood-testing equipment or assays is outside this research scope.

Pathologists at the centre of the research

Pathologists are essential research collaborators at every stage. Their expertise guides case selection, slide annotation, reference assessments, and the interpretation of model errors. They help distinguish clinically relevant findings from artefacts and assess the value of suggested follow-up investigations. AI outputs remain subject to their professional review.

Intended workflow

1. Sample results and slide images

   Prepare laboratory results across large batches of samples and digitised pathology slides for appropriately governed research analysis.

2. Data-quality checks

   Check completeness, measurement units, reference ranges, and slide-image quality before interpreting patterns.

3. Biomarker and image-pattern analysis

   Investigate statistical, machine-learning, and computer-vision methods for identifying patterns in laboratory results and slide images that may warrant closer review.

4. Flags with supporting rationale

   Present the laboratory measurements or slide-image regions underlying each flag and any suggested additional markers or parameters for consideration.

5. Pathologist review

   Have a pathologist assess the flag in the available clinical context and determine whether further investigation is appropriate.

Research governance

Use of patient-derived laboratory results and pathology slide images requires appropriate data-access permissions, privacy safeguards, and any required ethics or institutional approvals. Study protocols must define data handling, access controls, and the use of de-identified data where appropriate.

04 / Computer science & AI

From laboratory data to interpretable healthcare AI

The computer science component focuses on how large datasets of sample results and pathology slide images are represented, processed, and analysed. The AI component investigates relationships across biomarkers and image patterns in slides, with findings presented for pathologist assessment. The following areas define the research agenda.

Intended research outputs

The work aims to produce reproducible analysis pipelines, comparative model evaluations, and a prototype interface for pathologist review. These are research objectives; no deployed clinical AI service or validated diagnostic capability is claimed.

05 / Evaluation

Evaluation priorities

Evaluation will examine agreement with expert review, missed findings, false alerts, and the relevance of suggested follow-up investigations. Further evaluation is intended to assess performance across patient groups and laboratory settings.

Model evaluation will use separate development and test datasets, with sample records and slide images from the same patient kept within a single partition to reduce data leakage. Planned comparisons include rule-based baselines, sensitivity to missed review-worthy findings, false-alert burden, and performance across relevant patient groups. Evaluation of the review workflow will consider how pathologists use the outputs, alongside model performance.

Methodological reference: Good Machine Learning Practice guiding principles ↗

The research programme is in progress. These evaluation priorities guide the work; completed validation results and demonstrated clinical benefits are not reported on this page.

06 / Research leadership

Principal investigator

07 / Published work

Related research

08 / Collaboration

Research collaboration

We welcome conversations with pathologists, laboratory medicine researchers, computer scientists, machine-learning researchers, and health-data scientists interested in shaping clinically relevant research questions and evaluation methods.