ideasIdea
Read muscle loss automatically from scans patients already have
Every staging scan contains a precise measure of muscle mass that nobody looks at. Software could report it automatically and flag patients heading for wasting.
Automated segmentation of skeletal muscle at the third lumbar vertebra is accurate and fast, and low muscle mass predicts chemotherapy toxicity, surgical complications and survival across tumour types. The measurement is free because the scans already exist; the missing pieces are automated reporting into the record and a defined action when the value is low.
Hypothesis
Automated muscle metrics reported with every staging scan identify high-risk patients earlier than weight loss criteria, and triggering a supportive care pathway on that flag improves treatment completion rates.
Rationale
Opportunistic imaging biomarkers have already been adopted for bone density and coronary calcium from routine scans. Cachexia is currently diagnosed late, by weight loss that has already occurred, when reversal is hardest.
What would test it
Deploy automated segmentation in one centre and randomise by clinic to flag-triggered referral versus usual care; endpoints are treatment completion, dose intensity and hospital admissions.
Maturity
preclinical evidence
Who has to act
data
Cost to try
Small (under $1M)
Years to first evidence
3
Bottlenecks it attacks
- Cachexia, toxicity and the limits of the patient · Patients often die of wasting or cannot tolerate the doses that would work. Treating the patient, not just the tumour, lags far behind.
- AI that is built but not validated or deployed · Thousands of cancer AI models are published; a handful are in clinical use, and fewer have shown they help patients.
- Toxicity and quality of life are undervalued · Trials measure how long people live, not how they live. Side-effects are under-reported and under-treated.