Mingui Sun - US grants

Recent technological advances in electronic systems and computer applications have allowed over one-hundred channels of electroencephalograms (EEGs) or event-related potentials (ERPs) to be acquired simultaneously. By analyzing the recorded data using modern signal processing techniques, new insights into the functional activity of the brain have been provided. Despite these advances, tedious and primitive operations are required to affix a large number of electrodes on the scalp, making the high-resolution EEGs and ERPs unattractive to many hospitals and research laboratories. To solve this bottleneck problem, we propose an EEG electrode placement system based on novel engineering designs of electronic and mechanical components. This system automatically cleans the scalp, scratches the epidermis layer, drives an electrode array through the hair to the scalp sites, and presses the electrodes firmly in place during data acquisition. This system is also equipped with a set of sensors which allow automatic calculations of all scalp locations. In this Phase I effort we will prototype the critical components of the system and thoroughly test them. In future efforts we will construct a complete system that is capable of affixing over 100 electrodes on the scalp and measuring their coordinates within minutes.

The goal of this research is to develop an automatic machine to apply an array of electrodes to the human scalp to record electroencephalograms (EEGs). The EEG is widely utilized method to evaluate neurological function. Recent advances in computer and information systems have greatly enhanced high resolution EEG technology which permits recording, transmitting, storing, and analyzing hundreds of channels of data from densely located electrodes to map the functional activity of the brain. However, the procedures for affixing EEG electrodes on the human scalp have experienced little improvement. The required tedious and primitive manual operations have been a well known problem in the EEG community This problem is severely limiting the wide acceptance of the high-resolution EEG technology in clinical applications. We aim to solve this problem by designing new types of electrodes and developing an automatic system to affix electrodes to the scalp. In our Phase I STTR we have constructed several key mechanical and electronic components, and demonstrated the feasibility of our approach. We propose to continue this innovation in Phase II R&D. Our primary objectives are to construct a complete prototype system and to evaluate its performance. PROPOSED COMMERCIAL APPLICATION: The high costs of labor and slow turn-over of equipment due to EEG electrode manipulation are immense in the operation of clinical and research EEG laboratories. Therefore, the low-cost automatic EEG electrode placement system is expected to be well received by the EEG community. as a highly marketable commercial product.

DESCRIPTION (provided by applicant): Neurophysiological InatraOperative Monitoring (IOM) is an important procedure to reduce morbidity due to unexpected conditions developed during surgical manipulation. This procedure includes monitoring high-resolution, close-up video of the surgical field, analyzing simultaneously recorded neurophysiological data, and interacting with the neurosurgeon in cases where operative strategies need to be modified to avoid postoperative neurological deficits. IOM is traditionally performed within the operating room in medical centers where experienced neurophysiologists are available. In regional hospitals this monitoring is often difficult to perform because of the lack of experts. Over the years we have been studying multimedia remote IOM systems capable of transmitting neurophysiological data, voice, and images through the Interact, However, transmission of digital video with an acceptable quality for IOM is still a problem, even when a broad band Internet connection is utilized. The primary reason is the lack of a high-performance video compression algorithm adapting to the special features of IOM video. We will form a research team consisting of experienced electrical engineers, neurophysiologists, and neurosurgeons to attack this problem. We will develop a special-purpose video compression algorithms that minimize the bandwidth requirement. We will also develop new methods for synchronization and integration of the multimedia data for remote IOM. We will construct a prototype system capable of providing high-quality IOM service in rural regions within a 200-mile radius of the University of Pittsburgh Medical Center.

DESCRIPTION (provided by applicant): Obesity is associated with a large variety of genetic and environmental factors. In order to understand the etiology of this condition and develop effective weight management programs, accurate acquisition of diet and physical activity data in free-living environment is essential. Currently, self-reporting is the primary method for data acquisition, but does not accurately reflect the true habitual behavior of individuals in real life. As a result, the lack of assessment tools to produce unbiased, objective data has significantly hampered the progress of obesity research. We propose a novel application of multimedia technology to study obesity. It will be based on electronic chronicle (or e-chronicle), a powerful multimedia data management technology which provides an easily accessible electronic memory of individual's experience and daily events. Specifically, we will develop a unified sensor device which is cosmetically pleasant and can be easily worn by patients. This device will consist of a set of physiological sensors and a miniature video camera. The data recorded will be uploaded to a powerful computer where extensive multimedia processing will be performed to remove human appearances in the video and organize information using the e-chronicle technology. Diet and activity related events will be automatically extracted, indexed, and organized into an easily accessible form, providing a new platform technology to study lifestyle, behavior, and environment that promises new understanding and effective treatment option to manage obesity. We propose a novel application of multimedia technology based on electronic chronicle to study obesity. We will develop a unified sensor device consisting of a set of physiological sensors and a miniature video camera to acquire field data. Diet and activity related events will be automatically extracted, indexed, and organized, providing a new platform technology to study lifestyle, behavior, and environment that promises new understanding and effective treatment option to manage obesity.

DESCRIPTION (provided by applicant): Wearable eButton for Evaluation of Energy Balance with Environmental Context and Behavior Overweight and obesity have become a wide spread epidemic affecting more than 60% of the population in the United States. A recent study indicates that the estimated direct and indirect costs of obesity to the U.S. economy are at least $215 billion annually. In the battle against obesity and related diseases, both the research and clinical communities require an accurate tool to measure people's energy intake and expenditure in real life. However, presently, the most utilized tool is still a questionnaire which is subjective and often inaccurate. The current state-of-the-art in diet and physical activity measurement has fallen far behind what the modern technology is able to provide. The field is mature for a substantial innovation. In this application, we propose the development of a new electronic device which has the potential to produce a technological quantum leap in the measurement of diet and physical activity. This button- like device, eButton, will be worn naturally on the chest using a pair of magnets or a pin. A new measurement concept based on the use of the wearable computer will be utilized in our device design. The eButton will contain a low-power, high-performance microprocessor running a simplified version of the LINUX operating system. It will contain numerous innovative designs, including an optical eating detector to monitor eating/drinking/smoking, two miniature cameras that produce a stereo vision to measure food portion size without depending on a reference card, an ear-based oximeter for measurement of heart rate and oxygen saturation, and an extrapolation formula to measure outdoor environment using the US environmental protection agency (EPA) database. eButton will store the multimedia data acquired by a variety of advanced miniature sensors in a flash memory within the device. It will also have a wireless link to a smart phone which will allow researchers to monitor the operating status of eButton remotely in real time. All sensors and function modules will be individually controlled by software to allow researchers to select among the available system resources for their particular needs. Despite the wide functionality of the device, eButton requires an extremely low user respondent burden. The research participant is required to do nothing more than turning on/off the device and recharging its battery at night. During our research, eButton and associated algorithms/software will be designed and constructed in our laboratory by an experienced team of electronic/software engineers based on an early version of the device developed under the NIH GEI diet and physical activity research program. Once eButton is constructed, we will implement a thorough validation process using human subjects to evaluate its accuracy in diet and physical activity assessment by comparing against the doubly-labeled water method as the gold standard.

A Human-Mimetic AI System for Automatic, Passive and Objective Dietary Assessment Unhealthy diet is strongly linked to risks of chronic diseases, such as cardiovascular diseases, diabetes and certain types of cancer. The Global Burden of Disease Study has found that, among the top 17 risk factors, poor diet is overwhelmingly the No. 1 risk factor for human diseases. Despite the strong connection between diet and health, unhealthy foods with large portion sizes are widely consumed. Currently, 68.5% of U.S. adults are overweight, among the highest in developed countries. The recent decline in U.S. life expectancy sent another alarming signal about the general health of the American people. Understanding how the diet-related risk factors affect people?s health and finding effective ways to empower them in improving lifestyle habits are among the most important tasks in public health. Unfortunately, dietary assessment in real-world settings has been exceedingly complex and inaccurate to implement. Technology is needed that allows researchers to assess dietary intake easily and accurately in real world settings so that effective intervention to manage obesity and related chronic diseases can be developed. We propose a biomedical engineering project to address the dietary assessment problem, taking advantage of advanced mathematical modeling, wearable electronics and artificial intelligence. Our research team has been improving the ability to assess diet for over a decade. We have designed the eButton, a small wearable device pinned on clothes in front of the chest, capable of collecting image-based dietary data objectively and passively (i.e., without depending on subject?s self-report or volitional operation of the device). We have also developed algorithms to compute food volumes and nutrients from images. Since the eButton was developed, it has been used by many researchers in the U.S. and other countries for objective and passive diet-intake studies in both adults and children. Despite the past successes, there have been two lingering critical problems associated with the objective and passive dietary assessment using wearable devices: 1) substantial manual efforts are required for researchers to visually examine image data to identify foods and estimate their volumes (portion sizes), and 2) there are privacy concerns about researchers? viewing of participants? real-life images. Although solving these problems could enable the eButton and other wearable devices for large-scale diet-intake studies, we were not able to find effective solutions until recently when Artificial intelligence (AI) emerged. Advanced AI systems, especially those based on deep learning, can be trained by large amounts of labeled data to produce results comparable or even superior to those produced by human in numerous fields of applications. AI technology is also a powerful tool for dietary assessment, potentially providing an ideal solution to the two previously mentioned problems. We thus propose to develop a human-mimetic AI system to recognize foods from images, estimate portion sizes, and find energy and nutrient values from a database in a fully automatic process. Using the AI approach, there will be no need for researchers to view participants? real-life images, and the AI system well-respects individuals? privacy because it is trained to recognizes human foods only, nothing else. Currently, the performances of existing AI systems are limited by the extensive variety and high variability of human foods, insufficient training data, and difficulty in finding appropriate nutritional information from food databases. In this application, we propose a new strategy to personalize the AI system for each research participant using an advanced mathematical model of personal food choices. With this personalization step, the dimensionality of our envisioned AI system can be reduced drastically, and our goal of automatic, objective and passive dietary assessment can be reached realistically. We also propose to improve the electronic hardware and develop a biomimetic camera to enlarge the field of view for the eButton. Finally, we will conduct a thorough evaluation of the personalized AI system in real-world settings using human subjects.