Back

Using Independent Components Analysis To Identify Breast Cancer Using Dynamic Active Thermography

Gershenson, M.; Gershenson, J. P.

2025-11-25 oncology
10.1101/2025.11.23.25340812 medRxiv
Show abstract

Mammography is the gold standard for breast cancer detection, but there remains a need for supplementary techniques to find lesions it misses. This research explores dynamic farinfrared thermography, which uses a thermal camera to monitor changes in skin temperature over time. The study aims to characterize the thermal behavior of the vascular system, both at rest and in response to a temperature stimulus that causes Vaso modulation (blood vessel constriction or dilation). We use matrix factorization to reduce 20 thermal frames and the static frame into a final set of three frames and a time-trace matrix. The results yielded clear images of these thermal and vascular responses. Using the time matrix without classification into sick/healthy, I produce an algorithm with output of three parameters. With apparent clustering correlated with the patient classification as sick or healthy. The authors recommend further studies with more extensive clinical data to validate this new approach.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.