The objective of the research is to enable at-home care of patients with a prior myocardial infarction (MI) by increasing the efficacy of cardiovascular monitoring for mental stress. Towards this goal, we have completed work demonstrating that a novel multimodal wearable sternal patch can collect data capturing stress-induced physiological changes in the MI population. To elucidate the stress “signature” that is produced and ensure that it can be captured consistently, it is necessary to better understand the signals from which it is derived.  Towards this goal, we have completed work assessing the agreement of the one-lead electrocardiogram (ECG) patch features to those captured from a three-lead ECG. We have also completed preliminary qualitative work assessing the signal quality of three wavelengths of patch-derived photoplethysmogram (PPG) collected from the torso and elucidating the physiological differences that arise in PPG amplitude (PPGamp) as compared to that collected from the periphery. In our proposed work, we plan to develop improved methods to assess patch seismocardiogram (SCG) signal quality through the use of tools from topology. The final step is to assess how patch-derived signals and features are predictive of clinical metrics indicative of patient outcomes. We aim to develop novel algorithms using traditional signal processing, machine learning, and tools from topology to assess whether patch signals can be used for risk-stratification in this population. Such a contribution could enable the administration of crucial preventative interventions.