Monitoring real-world driver behavior for classification and early prediciton of Alzheimer's disease PI: Matthew Rizzo MD Investigators: Jeff Dawson, ScD, Shauna Hallmark, PhD, Jennifer Merickel, PhD, Dan Murman, MD, Soumik Sarkar, PhD, Anuj Sharma, PhD, Ying Zhang, PhD Funding: National Institute on Aging This project tackles grand challenge of using a person’s own vehicle as a passive-detection system for flagging potential age- and/or disease-related aberrant driving that may signal early warning signs of functional decline and incipient Alzheimer’s disease (AD), even before standard clinical tests. The proposal builds on current successes using sensor technologies and a taxonomy of driver errors to comprehensively characterize patterns of real-world older driver behavior, exposure, and safety over extended time frames. These ongoing achievements support an extraordinary opportunity in this research project to use a patient’s own vehicle and devices as diagnostic tools to rigorously classify and track heretofore inaccessible fingerprints of driver behavior, sleep, and mobility––in the real world––to develop models capable of early detection and prediction of functional decline in older patients at risk for AD.
Safety and efficacy of cannabidiol treatment in patients with PTSD PI: Matthew Rizzo, MD Investigators: Soonjo Hwang, MD, Jennifer Merickel, PhD, Brigette Soltis-Vaughan, MSN, APRN-BC, NP, Ying Zhang, PhD, Funding: UNMC Foundation The objective of this proposal is to determine the safety and efficacy of cannabidiol (CBD) for treatment of patients with Post-Traumatic Stress Disorder (PTSD). Clinical interest in applications of cannabis plant products for the treatment of various mental health disorders has risen significantly in the past decade. CBD has been suggested for the treatment of psychopathologies associated with serotonergic dysfunction including mood and anxiety disorders and PTSD. This study is comprehensively evaluating the efficacy of CBD in treating PTSD, including effects on clinical symptoms, neurobiology, inflammation and real-world patient outcomes through objective measurements of patient safety, mobility, quality of life and sleep.
Deep Insight: Deep Extraction of Driver State from Naturalistic Driving Dataset PI: Anuj Sharma, PhD Investigators: Yaw Adu-Gyamfi, PhD, Jennifer Merickel, PhD, Matthew Rizzo, MD, Soumik Sarkar, PhD, Senem Velipasalar, PhD Funding: Federal Highway Administration Computer vision algorithms have been improved significantly to provide accurate detection, tracking and recognition of objects in general. Available public datasets are crucial for extracting key features thanks to ever developing machine learning and deep learning algorithms in general. Since cameras are becoming vastly available to be employed on vehicles to extract useful information for both autonomous driving and intelligent driver assistance, we are targeting to develop intelligent driver state estimation algorithms that are based on state-of-the-art detection and recognition using computer vision. One of the main drawbacks for naturalistic driving data is having low-resolution and noisy video data that limits the overall accuracy when we test with the models trained on clear images. We are proposing (i) comprehensive AI platform for data management, modeling and enhanced annotations, (ii) video quality enhancement using deep models, iii) face detection at acute angles, and (iv) Recurring network-based driver state estimation.
Driving Performance and Safety in Rheumatoid Arthritis PI: Ted Mikuls, MD Investigators: Kaleb Michaud, PhD, Jennifer Merickel, PhD, Matthew Rizzo, MD Funding: Rheumatology Research Foundation Automobile driving is an instrumental activity of daily living and is linked to health, well-being, and quality of life in patients with Rheumatoid Arthritis (RA). Despite its importance, there is little prior research investigating driving risk in patients with RA. To address this knowledge gap, we are using state-of-the-art simulation and naturalistic driving assessment technologies to link clinical metrics of RA disease severity to real-world driver behavior and mobility. Results from this study will fill critical gaps in RA patient care and develop "toolkits" to preserve mobility and quality of life for patients with RA.
Driving safety in fluctuating Parkinson’s disease PI: Jennifer Merickel, PhD Investigators: John Bertoni, MD, Danish Bhatti, MD, Matthew Rizzo, MD, Ergun Uc, MD Funding: UNMC Foundation Parkinson's Disease (PD) is a relatively common neurodegenerative disorder affecting movement, vision, cognition, sleep, and mood. These impairments, in turn, can affect the performance and behavior of a patient with PD in safety-critical tasks such as automobile driving. This pilot research study addresses real-world driving and risk in drivers with and without PD using in-vehicle sensors and activity monitors. The results can help us to better understand how driving safety in PD depends on a driver's disease, medication dosage, and functional abilities. The findings will ultimately help improve clinical recommendations, safety, mobility, and quality of life in patients with PD.
Quantitative assessment of driving capabilities of senior drivers PI: Anuj Sharma, PhD and Jennifer Merickel, PhD Funding: Toyota Collaborative Safety Research Center This project will develop a driver performance evaluation system using real-world driving data that can be used to provide older drivers and their families feedback on real-world driving risk. To accomplish this goal, key performance metrics quantifying older driver safety will be identified and used to establish a valid scoring system to rate older driver performance. Developed models will be used to build a smartphone application to provide older drivers and their families real-time feedback on driving risk to preserve safety mobility in aging.
Developing mechanisms to detect driver risk from driver state in diabetes PI: Anuj Sharma, PhD and Jennifer Merickel, PhD Investigators: Matthew Rizzo, MD and Soumik Sarkar, PhD. Funding: Toyota Collaborative Safety Research CenterThis project will advance mechanistic models capable of predicting acute, in-vehicle medical emergencies to develop advanced driver assistance systems that support safe mobility in drivers with diabetes. To accomplish this goal, big data analytic techniques will be used to mine vehicle video and sensor data to detect driver state changes that correlate with acute, in-vehicle glycemic episodes in drivers with type 1 and type 2 diabetes. This project will provide insight into driver state features and their interdependencies that can be quantified to probabilistically index acute glycemic changes before and during events.