Related Experiment Video
Updated: Aug 13, 2025
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
6.9K
Using Machine Vision of Glycolytic Elements to Predict Breast Cancer Recurrences: Design and Implementation.
1Department of Ophthalmology and Visual Sciences, University of Michigan Medical School, 1000 Wall Street, Ann Arbor, MI 48105, USA.
Metabolites
|January 21, 2023
Summary
This study introduces a machine vision test using glycolytic patterns in tissue samples to predict cancer recurrence risk from ductal carcinoma in situ (DCIS) lesions. The method accurately identifies early metabolic changes, aiding in early cancer recurrence detection.
Area of Science:
- Biomedical research
- Cancer biology
- Metabolic reprogramming
Background:
- Early identification of cancer recurrence is crucial for patient outcomes.
- Ductal carcinoma in situ (DCIS) can precede invasive cancer recurrence.
- Metabolic alterations, particularly glycolysis, are implicated in cancer progression.
Purpose of the Study:
- To evaluate the efficacy of a machine vision test in identifying glycolytic patterns within DCIS lesions.
- To determine if these patterns can predict subsequent cancer recurrences.
- To explore the mechanistic role of metabolic reprogramming in early cancer development.
Main Methods:
- Utilized conventional formalin-fixed paraffin-embedded tissue samples.
- Employed machine vision to analyze glycolytic enzyme and transporter patterns (phospho-Ser226-GLUT1, PFKL).
- Stratified patient micrographs based on the probability of originating from a recurrent sample.
Main Results:
- The machine vision test accurately identified subpopulations of cells preceding cancer recurrence.
- Stratification removed overlap between recurrent and non-recurrent patients, eliminating false positives/negatives.
- Computationally positive samples indicated recurrence, while negative samples indicated non-recurrence.
Conclusions:
- Glycolytic enzyme and transporter patterns in DCIS lesions can predict cancer recurrence.
- Machine vision analysis of these patterns offers a quantitative and accurate method for risk stratification.
- Understanding early metabolic reprogramming provides insights for novel therapeutic strategies.
