Abstract

Clinical trials of coherence-gated Doppler infrared spectroscopy of intracellular dynamics in living tumor tissue seek to identify the efficacy of prescribed chemotherapy for cancer patients. Changes in intracellular dynamics have specific Doppler signatures that depend on the applied cancer drugs and the sensitivity of the patient to treatment. A challenging feature of these assays is a strong intra-tumor heterogeneity that poses a significant challenge to machine-learning classifiers. We train a Twin Deep Network (TDN) to identify these signatures in the presence of strong heterogeneous background to accurately predict patient response to therapy. The TDN is being applied to two ongoing clinical trials: a clinical trial of HER2neg breast-cancer patients and esophageal cancer patients, all undergoing neoadjuvant therapy or chemoradiation therapy, respectively. This work provides insight into the value of Deep Learning for advanced data analytics as the volume and variety of data from optics-based assays grows.

© 2021 The Author(s)