Medical science concept based artificial intelligence to identify the blood vessels in kidney

Development of medical science concept based artificial intelligence algorithm and comparison with conventional artificial intelligence algorithm for detection of blood vessels in histological slides - NIL

Status

Active, not recruiting

Phases

Phase 1

Study type

Interventional

Source

CTRI

Registry ID

CTRI/2025/04/083902

Enrollment

500

Registered

2025-04-02

Start date

Unknown

Completion date

Unknown

Last updated

2025-04-28

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

None listed

Interventions

Intervention1: human intelligence-based medical science concepts: medical science concept (histological definition of artery, vein, arterioles etc) enrolled human intelligence-based steps will be defi

Eligibility

Inclusion criteria

Inclusion criteria: H & E stained histological images of kidneys for algorithm development will be used. Inclusion criteria: H & E stained histological images of kidneys annotated by experts for the ground truth will be included in the study. Images captured of any magnification will be included in the study.

Exclusion criteria

Exclusion criteria: 1. Images with artefacts, or improper labelling will not be included in the study. 2. Blurred images will be excluded from the study. 3. Histological slides from other than humans will not be included in the study. 4. Histological diagrams drawn by digital software will not be included in the study.

Design outcomes

Primary

Measure	Time frame
Ability to identify the blood vessels in the histological slides of the kidney. Models developed using medical science concept based artificial intelligence algorithm in comparison to conventional artificial intelligence algorithm.Timepoint: At baseline.	—

Secondary

Measure	Time frame
Comparative efficiency of human intelligence-based AI vs. conventional AI Timepoint: At the baseline;Generalizability & scalability of AI models. Ability to adapt to different histological images & maintain accuracy across datasetsTimepoint: after 10 weeks;Computational efficiency Time required for model inference & image processingTimepoint: At the Baseline	—

Countries

India

Contacts

Public ContactPRADIP R CHAUHAN

All India Institute of Medical Sciences, Rajkot

prajjawalitresearch@gmail.com08866199560

Outcome results