The cardiopulmonary system is responsible for the timely delivery of oxygenated blood to key sites throughout the body, requiring an interconnected network of cardiovascular, pulmonary, and autonomic mechanisms. These mechanisms include regulation of heart rate, blood pressure, and vascular tone alongside efficient ventilation and coordinated respiratory muscle function. Disruptions in this intricate system, such as autonomic dysfunction in Parkinson’s disease (PD) or progressive muscular degeneration in Duchenne’s Muscular Dystrophy (DMD), can extensively impair cardiopulmonary function. However, existing technologies for monitoring such impairments, particularly those related to respiratory monitoring, are often confined to clinical settings or provide only temporally and physiologically coarse indicators of cardiopulmonary status. To overcome these limitations, we developed multimodal sensing systems and machine learning techniques to enhance the assessment of cardiopulmonary health. We first evaluated the clinical and at-home utility of a multimodal chest patch in a population of patients with cardiovascular autonomic dysfunction. Then, as respiratory sensing technology is less established than cardiovascular sensing, we assessed the effectiveness of bioimpedance (BioZ)-based respiratory sensing for unobtrusive estimation of tidal volume and respiratory timings in static and ambulatory settings. Finally, based on these prior findings, we developed and validated the reSPIRE system, a multimodal chest patch that incorporates the prior measures of cardiovascular and respiratory physiology alongside signals related to respiratory muscle activity as a holistic approach to cardiopulmonary monitoring. The work presented herein illustrates an advancement in wearable technology by enabling unobtrusive measurements of physiologically diverse cardiopulmonary markers that can be leveraged for the early detection and treatment of cardiopulmonary impairments in real-world settings.