Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Unified Multi-task Learning for Colorectal Cancer Diagnosis via Uncertainty-aware Routing, Cross-task Consistency,

Divya Midhunchakkaravarthy¹, G Muni Nagamani², Lakshman Narayana V³

¹Department of Electronics and Communication Engineering, Lincoln University College, Petaling Jaya, 47301, Malaysia.

Current Pharmaceutical Biotechnology

|May 15, 2026

Summary

Related Concept Videos

Imaging Studies III: Gastrointestinal Motility Studies and Virtual Colonoscopy

Imaging Studies III: Gastrointestinal Motility Studies and Virtual Colonoscopy

This lesson explores three gastrointestinal imaging techniques: radionuclide testing, colonic transit studies, and virtual colonoscopy.
Radionuclide Testing
Radionuclide testing is a sophisticated medical technique for assessing gastrointestinal motility. It focuses on gastric emptying and colonic transit time. Radioactive markers track the movement of food through the digestive system, providing insights into gastrointestinal disorders.
In gastric emptying studies, a meal's liquid and solid...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Graph neural networks for networked analysis of gestational diabetes risk factors: a multi method framework.

Scientific reports·2026

Same author

Dynamic cross-domain transfer learning for driver fatigue monitoring: multi-modal sensor fusion with adaptive real-time personalizations.

Scientific reports·2025

Same author

Design of an improved graph-based model for real-time anomaly detection in healthcare using hybrid CNN-LSTM and federated learning.

Heliyon·2025

Same author

Towards Energy-Efficient and Delay-Optimized Opportunistic Routing in Underwater Acoustic Sensor Networks for IoUT Platforms: An Overview and New Suggestions.

Computational intelligence and neuroscience·2022

Same author

Industry 5.0: Ethereum blockchain technology based DApp smart contract.

Mathematical biosciences and engineering : MBE·2021

This summary is machine-generated.

This study introduces a unified AI framework for colorectal cancer diagnosis, improving tumor grading and tissue segmentation accuracy. The novel approach enhances diagnostic consistency and efficiency in histopathology, reducing annotation burden.

Area of Science:

Computational pathology
Artificial intelligence in oncology
Digital histopathology

Background:

Colorectal cancer (CRC) diagnosis relies on histopathology, often involving separate segmentation and grading tasks.
Independent task analysis leads to redundancy, increased annotation burden, and reduced model generalization.
Developing a unified framework for joint segmentation and grading is crucial for improving efficiency and accuracy.

Purpose of the Study:

To develop a unified multi-task learning framework for simultaneous tissue segmentation and tumor grading in colorectal cancer whole-slide images.
To enhance label efficiency, model robustness, and generalization while minimizing annotation requirements.
To improve diagnostic precision and model efficiency in histopathological analysis.

Main Methods:

Keywords:

Colorectal cancer histopathology multi-task learning process.tissue segmentation tumor grading

Related Experiment Videos

Hierarchical Uncertainty-Gated Task Routing (HUGTR) for dynamic feature allocation based on uncertainty.
Cross-Task Consistency Attention Matrix (CTCAM) to enforce spatial coherence between segmentation and grading.
Adaptive Label Denoising with Structural Priors (ALDSP) using graph convolutional autoencoders.
Contrastive Segmentation-Grading Latent Embedding (CSGLE) for aligning latent representation spaces.
Curriculum-Based Multi-Resolution Task Cascade (CMRTC) for progressive training from low to high resolutions.

Main Results:

Achieved a 6.8% increase in tumor grading AUC and a 3.5% improvement in segmentation Dice score.
Reduced model parameters by 27% and inter-observer variability by 14.3%.
Demonstrated enhanced diagnostic precision and model efficiency through integrated task learning.

Conclusions:

The unified multi-task framework sets a new benchmark for colorectal cancer histopathological analysis.
Innovations enable better annotation utilization, improved diagnostic consistency, and enhanced scalability.
The framework serves as a robust AI tool for pathology workflows, integrating segmentation and grading effectively.