Skip to content

A lab study using a dye glow camera and artificial intelligence to quickly check on the removed tissue sample whether cancer has been fully removed in breast and colorectal surgery.

Smart Surgical Margins: An Artificial Intelligence Powered RealTime Cancer Residual Detection Using Fluorescence Imaging System: A Cross-Sectional Study at a Tertiary Care Hospital in Mandya - NIL

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
CTRI
Registry ID
CTRI/2025/11/097943
Enrollment
120
Registered
2025-11-24
Start date
Unknown
Completion date
Unknown
Last updated
2025-12-08

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

Conditions

Health Condition 1: C50- Malignant neoplasm of breast Health Condition 2: C189- Malignant neoplasm of colon, unspecified Health Condition 3: C19- Malignant neoplasm of rectosigmoidjunction Health Condition 4: C20- Malignant neoplasm of rectum

Interventions

Intervention1: Nil: Nil Intervention2: Ex vivo ICG fluorescence imaging with computer vision and machine learning: Fresh inked surgical specimen is imaged in a light tight box under 785 nm excitation

Sponsors

Mandya Institute of Medical Science MIMS
Lead Sponsor
Indian Council for Medical Research ICMR
Collaborator
Manmed Dynamics LLP
Collaborator

Eligibility

Inclusion criteria

Inclusion criteria: Adults aged 18 years and above Primary breast cancer or colorectal cancer scheduled for surgery with curative intent R0 goal Fresh excised specimen available immediately for ex vivo imaging in the laboratory Indocyanine green used as per routine care or eligible for topical ex vivo protocol as approved by ethics Specimen orientation recorded and margins inked before imaging Time from excision to imaging within four hours Written informed consent obtained or ethics approved waiver for discarded tissue

Exclusion criteria

Exclusion criteria: Metastatic disease or palliative cytoreductive resections Known hypersensitivity to indocyanine green or iodides when systemic indocyanine green is used Pregnancy or lactation when systemic indocyanine green is not permitted by local policy Specimen unsuitable for imaging due to heavy cautery fragmentation inadequate fluorescence or gross contamination Time to imaging more than four hours after excision Uncontrolled diabetes as judged by the treating team if you wish to retain this from your proposal Any serious medical or logistical condition that prevents safe handling imaging or timely processing of the specimen Any exclusion mandated by the Institutional Ethics Committee

Design outcomes

Primary

MeasureTime frame
Per-face sensitivity and specificity of the index test (ex-vivo ICG fluorescence and Machine Learning) to detect positive or close margins compared with histopathology (H and E) as reference standard.Timepoint: Day 0 within 0 to 4 hours after surgical excision imaging performed before histopathology sampling

Secondary

MeasureTime frame
Area under the ROC curve (AUC) for face-level classification of margin status.Timepoint: Day 0 1;umour-to-Background Ratio (TBR) threshold optimizing the Youden index.Timepoint: Day 0 1;Time-to-result (minutes) from specimen arrival in the lab to margin heatmap.Timepoint: Day 0;Per-slice sensitivity, specificity, PPV/NPV versus histopathology.Timepoint: Day 0 1;Concordance between AI hotspot location and histologic margin distance ( m).Timepoint: Day 0-1

Countries

India

Contacts

Public ContactIng Vishwas Gowda P N

Mandya Institute of Medical Science

1983lingu@gmail.com9480387075

Outcome results

None listed

Source: CTRI (via WHO ICTRP) · Data processed: Feb 4, 2026