Samsung Research America (SRA) was founded in 1988, headquartered in Mountain View at the heart of Silicon Valley, with offices across the United States and Canada. SRA not only witnessed Samsung’s history in the Bay Area, but also it is at the forefront of cutting-edge technologies to create new businesses and developing core technology including artificial intelligence (AI), 5G/6G, digital health, enterprise security, mobile innovation, and so on. SRA also plays a key role in providing the infrastructure to support Samsung’s open innovation and university collaboration activities. SRA aims to create new businesses and developing core technology to enhance the competitiveness of Samsung products to impact the future.
Among various core technologies that SRA is focusing on, SRA’s Digital Health Lab is at the forefront of artificial intelligence (AI) and machine learning (ML) to develop clinically explainable digital biomarkers to assess various health conditions and their severity. Our talented and multi-disciplinary team develops innovative technologies and tools to aid clinicians and their patients in monitoring and managing conditions such as heart disease, lung disease, neurological disorders, diabetes, and sleep disorders, among others.
We utilize cutting-edge sensor technology to capture physiological responses and apply our advanced signal processing and machine learning frameworks to build algorithms that detect, predict disease conditions, and deliver interventions through digital devices such as smartphone, smartwatch, and earbuds. Our technologies facilitate remote patient monitoring, enabling patients to make behavioral changes that improve their health and quality of life.
My name is Mahbubur Rahman and I hold the position of Senior Staff Research Engineer in SRA Digital Health Lab. I am the lead researcher on multiple patient-centered research projects that focus on assessing and monitoring human health and behaviors. Specifically, I am responsible for developing and validating interpretable and comfortable signal processing and machine learning algorithms to monitor lung health, mental health, and cardio-respiratory fitness using smartphone, smartwatch, and earbuds. Additionally, I am leading the charge in translating novel health algorithms into world’s first, ground-breaking digital health products, with the aim of enhancing and expanding the digital health market.
Passive collection and analysis of data related to disease diagnosis, prevention, and management is possible with mobile health (mHealth) technologies that leverage sensors, mobile apps, and wearables. With mobile health, it is now possible to monitor and intervene with users anywhere and anytime. The proliferation of smartphone usage globally means that our novel AI/ML algorithms on mobile sensor data have the potential to impact millions of patients and healthy individuals in monitoring, managing, and improving their health and wellness.
For instance, our breathing monitoring algorithms utilize low-power, less privacy-sensitive accelerometers in smartphones, smartwatches, or earbuds to estimate comprehensive respiratory biomarkers like breathing rate, depth, and inhalation-exhalation ratio. Our machine learning algorithms on forced breathing sound captured by a smartphone microphone can estimate lung functions that previously required expensive and bulky spirometers available only in hospitals. These biomarkers, along with heart rate and oxygen saturation (SpO2) measures obtained using the same mobile devices, can aid in monitoring and managing asthma and chronic obstructive pulmonary disease (COPD), which is the third leading cause of death worldwide.
The ability to conveniently assess and monitor using the user’s own mobile devices enables and empowers users to take control of their health while providing clinicians with longitudinal data from patients for personalized and effective clinical decisions.
Our research projects have resulted in more than 65 peer-reviewed papers in top-tier venues and more than 25 patents, and received multiple external and Samsung internal awards.
Our paper on estimating lung function using breathing audio collected via a smartphone has received Industry Best Paper award at the Institute of Electrical and Electronics Engineers (IEEE) International Conference on Pervasive Computing and Communications (PerCom 2020). And another paper on smartphone based breathing motion and wheeze sound detection with machine learning algorithms is published in the Association for Computing Machinery (ACM) Conference on Human Factors in Computing Systems (CHI) 2020 which is the biggest and the most prestigious conference in human-computer interaction with Google Scholar h5-index 113.
Breathing rate monitoring in the wild is challenging since breathing motion and sound are very subtle and background noises oftentimes mask the breathing signal. We have devised a novel solution by utilizing the earbud motion and acoustic sensor for passive, low-power, comfortable breathing rate monitoring which has the potential to passively monitor chronic respiratory diseases such as COPD, asthma, and infectious respiratory diseases such as flu or COVID-19. Our paper on this algorithm is published in the ACM CHI 2023.
Our novel algorithm on fusing motion sensor data with acoustic data using teacher-student model enables smartphone to detect and analyze fine-grained breathing phases, and is published in the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT) 2021 journal, the most prestigious venue for mobile, wearable, and ubiquitous technology. This work has also won a Samsung internal best paper award.
Our algorithm utilizing machine learning on accelerometer and photoplethysmography (PPG) sensor data from earbuds can detect psycho-social stress which is published in the IEEE Engineering in Medicine and Biology Conference (EMBC) 2022. Earbud accelerometer based real-time, passive breathing biomarkers tracking to help users effectively perform mindful breathing exercises for stress relaxation is published in the ACM International Conference on Mobile Human-Computer Interaction (MobileHCI) 2022.
Moreover, natural speech based pulmonary patient detection has won best poster award in the IEEE Body Sensor Network (BSN) 2019, behavioral privacy preservation techniques in mobile sensor data won honorable mention in the ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) 2016, and conversion episode detection from respiratory timeseries data collected in field was nominated for the best paper award in the ACM Wireless Health 2011.
Furthermore, I am honored to have represented Samsung and served as the chair for the Consumer Technology Association (CTA) working group (WG) focused on standardizing consumer technology for respiratory monitoring. This WG includes other key players in the digital health tech market. Respiratory monitoring standard developed by our CTA WG has been published and will help reshape the development and validation of future respiratory monitoring technologies. Additionally, it has been rewarding to represent Samsung at the Health and Wellness Group in the Connectivity Standard Alliance (CSA), where we work to define simple, interoperable, reliable, and secure standards for smart-home and internet-of-things (IoT) devices.
Healthcare is experiencing a significant paradigm shift caused by multiple factors such as evolving patient demands, advancements in digital health technologies, and the impact of the Covid-19 pandemic. The shift is leading towards more personalized, predictive, preventive, and convenient care, which is made possible by lowering the barriers to healthcare through telehealth and tech-enabled home care.
As patients get more used to receiving data-driven and personalized care, they have a greater expectation for more individualized healthcare services. SRA Digital Health Lab is ideally positioned to meet this demand by using a consumer-first approach and leveraging data to provide accessible and holistic care.
Our focus is on developing digital health technologies that can transform the healthcare system from being reactive and centered around curing diseases to a proactive approach that emphasizes prevention. This shift will benefit patients, physicians, and the overall sustainability of the healthcare system.
Using data-driven algorithms, we aim to provide patients with valuable insights that empower them to monitor their health conditions and enable longitudinal data collection. Our efforts will create a virtuous cycle that drives the development of digital health technologies. Our approach will increase efficiency, predictive capabilities, and personalization in healthcare. We aspire to inspire other stakeholders to expand the horizon of digital health technology and democratize healthcare access for all socio-economic statuses across the world.