Automatic detection of pain using machine learning.

Pain is one of the most common symptoms reported by individuals presenting to hospitals and clinics and is associated with significant disability and economic impacts; however, the ability to quantify and monitor pain is modest and typically accomplished through subjective self-report. Since pain is associated with stereotypical physiological alterations, there is potential for non-invasive, objective pain measurements through biosensors coupled with machine learning algorithms. In the current study, a physiological dataset associated with acute pain induction in healthy adults was leveraged to develop an algorithm capable of detecting pain in real-time and in natural field environments. Forty-one human subjects were exposed to acute pain through the cold pressor test while being monitored using electrocardiography. A series of respiratory and heart rate variability features in the time, frequency, and nonlinear domains were calculated and used to develop logistic regression classifiers of pain for two scenarios: (1) laboratory/clinical use with an F1 score of 81.9% and (2) field/ambulatory use with an F1 score of 79.4%. The resulting pain algorithms could be leveraged to quantify acute pain using data from a range of sources, such as ECG data in clinical settings or pulse plethysmography data in a growing number of consumer wearables. Given the high prevalence of pain worldwide and the lack of objective methods to quantify it, this approach has the potential to identify and better mitigate individual pain.

Automated stress detection using mobile application and wearable sensors improves symptoms of mental health disorders in military personnel.

Leading causes in global health-related burden include stress, depression, anger, fatigue, insomnia, substance abuse, and increased suicidality. While all individuals are at risk, certain career fields such as military service are at an elevated risk. Cognitive behavioral therapy (CBT) is highly effective at treating mental health disorders but suffers from low compliance and high dropout rates in military environments. The current study conducted a randomized controlled trial with military personnel to assess outcomes for an asymptomatic group (n = 10) not receiving mental health treatment, a symptomatic group (n = 10) using a mHealth application capable of monitoring physiological stress via a commercial wearable alerting users to the presence of stress, guiding them through stress reduction techniques, and communicating information to providers, and a symptomatic control group (n = 10) of military personnel undergoing CBT. Fifty percent of symptomatic controls dropped out of CBT early and the group maintained baseline symptoms. In contrast, those who used the mHealth application completed therapy and showed a significant reduction in symptoms of depression, anxiety, stress, and anger. The results from this study demonstrate the feasibility of pairing data-driven mobile applications with CBT in vulnerable populations, leading to an improvement in therapy compliance and a reduction in symptoms compared to CBT treatment alone. Future work is focused on the inclusion of passive sensing modalities and the integration of additional data sources to provide better insights and inform clinical decisions to improve personalized support.

Tissue Oxygenation Changes to Assess Healing in Venous Leg Ulcers Using Near-Infrared Optical Imaging.

Objective: Venous leg ulcers (VLUs) comprise 80% of leg ulcers. One of the key parameters that can promote healing of VLUs is tissue oxygenation. To date, clinicians have employed visual inspection of the wound site to determine the healing progression of a wound. Clinicians measure the wound size and check for epithelialization. Imaging for tissue oxygenation changes surrounding the wounds can objectively complement the subjective visual inspection approach. Herein, a handheld noncontact near-infrared optical scanner (NIROS) was developed to measure tissue oxygenation of VLUs during weeks of treatment. Approach: Continuous-wave-based diffuse reflectance measurements were processed using Modified Beer-Lambert's law to obtain changes in tissue oxygenation (in terms of oxy-, deoxy-, total hemoglobin, and oxygen saturation). The tissue oxygenation contrast obtained between the wound and surrounding tissue was longitudinally mapped across weeks of treatment of four VLUs (healing and nonhealing cases). Results: It was observed that wound to background tissue oxygenation contrasts in healing wounds diminished and/or stabilized, whereas in the nonhealing wounds it did not. In addition, in a very slow-healing wound, wound to background tissue oxygenation contrasts fluctuated and did not converge. Innovation: Near-infrared imaging of wounds to assess healing or nonhealing of VLUs from tissue oxygenation changes using a noncontact, handheld, and low-cost imager has been demonstrated for the first time. Conclusion: The tissue oxygenation changes in wound with respect to the surrounding tissue can provide an objective subclinical physiological assessment of VLUs during their treatment, along with the gold-standard visual clinical assessment.