National Seminar on Computer Vision & Image Processing

Akbar Abbas Doctor

Title: Test Automation-Medical Devices

Abstract: Quality Management and Control is an important aspect for any product development and manufacturing. Quality should be a part of design and manufacturing which inevitably improves the customer experience. To achieve high quality standards companies around the world deal with internal and external costs. External costs of quality are always a bad idea while internal costs should be carefully planned and implemented. Inspection and testing are the 2 important tools of internal quality control, aka. cost of conformance. For medical devices this cost of conformance is of paramount importance as external costs of quality is not mere loss of business or liability, but it is loss of human life.

Hence, inspection and testing becomes an integral and one of the most important process on the production floor for medical devices. Most companies thrive for automation in testing to get rid of human error. This drive for automation leads to implementation of all facets of technology to develop a seamless and errorless process. RT, Vision, cloud computing, robotics, statistical analysis all come together to create a process which delivers a safe product to the patients and filters the fall-outs which invariably saves lives.

Bio Sketch: Akbar Abbas Doctor
Principal Test Engineer-Fresenius Medical Care.
Leading Test Automation team for Product transfer from Fresenius-Germany to Fresenius-North America
Current residence: Concord, California. (San Francisco-Bay area)
Bachelors of Engineering from SCET, Surat, Gujarat.
Master in Biomedical Engineering from Furtwangen University, Germany

Chang-Hyun Oh

Title: Real-Time interactive MR Thermometry using Deep Neural Network

Abstract: Real-time interactive magnetic resonance (MR) temperature mapping of aqueous and adipose tissue is a challenging technique for MR-guided thermal therapy systems. To acquire the real-time interactive temperature map, the problems of long acquisition during MRI scans and post-processing time must be addressed. There are three major considerations when it comes to reducing the acquisition and processing times. First, the temperature of adipose and aqueous tissue must be calculated in a real-time manner; temperature images for adipose and aqueous tissue are calculated by T1 relaxation time and proton resonance frequency-based MR thermometry, respectively. Second, the acquisition time during the MRI scans should be reduced. Third, the time for processing the low-resolution MR images to high-resolution MR images and calculating the T1 relaxation map should be reduced. To overcome these major challenges, a cascaded convolutional neural network framework is proposed by utilizing the optimized MR thermometry protocols which enables real-time MR temperature images reconstruction simultaneously during MRI scans.

Bio Sketch: Chang-Hyun Oh is a full-time Professor in the Department of Electronics and Information Engineering, Korea University (KU) and principal researcher in the Medical Imaging System & Science Laboratory (MISSL), KU, South Korea. Before joining KU in 1993, he worked as a Research Scientist from 1985 to 1987, and Assistant Professor from 1987 to 1992 at Columbia University, New York, USA. Currently, He is also a member of Advisory Committee, System and Control Society, Korea Electronics Engineers Association, and the president of MRI Study Group, Korea Smart Medical Device Industry Promotion Foundation and Medical Imaging System Research Study Group, Korea Institute of Electronics Engineers. He was awarded the Grand prize for his contribution in the MRI field by the Korean Society for Magnetic Resonance in Medicine in 2018.His major research field is related to the development of Magnetic Resonance Imaging (MRI) – both hardware and software development. He has finished number of projects and current research projects include MRI guided (MRg) therapy such as MR-HIFU, MR-Hyperthermia, MRI Hardware and System development for low (<3T) and high (>3T) field MRI systems, Advanced MR Sequence Development for quantitative and qualitative applications, etc.

Chintan Parmar

Title: AI applications for medical imaging biomarkers

Abstract: Artificial intelligence (AI) algorithms, particularly deep learning, have demonstrated remarkable progress in image-recognition tasks. Methods ranging from convolutional neural networks to variational autoencoders have found myriad applications in the medical image analysis field, propelling it forward at a rapid pace. Historically, in biomedical fields, trained experts/physicians visually assessed medical images for the detection, characterization and monitoring of diseases. AI methods excel at automatically recognizing complex patterns in imaging data and providing quantitative, rather than qualitative, assessments. This talk will focus on AI applications for medical imaging based biomarkers.

Bio Sketch: Dr. Parmar is a Senior Expert in Oncology Data Science at Novartis Institutes for Biomedical Research, Cambridge, USA. His research interests in biomedical informatics spans many application domains, from data science, to processing and mining biomedical data, to advancing the state of statistical models in healthcare. In particular, mining and interpreting complex and multimodal biomedical data to decode novel characteristics of different diseases and enhance the clinical decision support. Prior to his current role, Dr. Parmar was a postdoctoral researcher at Harvard Medical School, USA and graduate research scholar at Maastricht University, The Netherlands and Indian Statistical Institute-Kolkata, India. He has published several articles in these research areas.

Mukta Joshi

Title: Advances in medical imaging for diagnosis and treatment

Abstract: Healthcare systems today need to deliver better care to more people at lower cost. This is the main driving force for innovation in healthcare technology. Imaging plays a major role in gathering information about conditions non-invasively as well as enabling image guided procedures. In order to deliver better care, faster and at a lower cost, imaging needs to provide definitive diagnoses and improve the accuracy and efficiency of image-guided procedures.

In this talk, we look at the current healthcare landscape and learn about specific technologies in various imaging modalities such as computed tomography (CT) and Image Guided Therapy (IGT) and their impact on clinical practice.

Bio Sketch: I have over 22 years’ experience in the field of medical imaging. I have been involved in multiple stages of the product development lifecycle from road mapping to research to clinical adoption. I am an engineer with a passion to bring meaningful solutions that address clinical problems. I have led multiple projects with increasing level of responsibilities working at both GE healthcare and Philips Healthcare. These have spanned various imaging modality including CT, Ultrasound and Image Guided Therapy bringing innovations to clinical practice. I hold a Master of Science degree in Electrical Engineering in Brain imaging and have authored several patents.

Ram Bilas Pachori

Title: Detection of human brain disorders using novel machine learning approaches

Abstract: In this talk, the motivation for computer-aided diagnosis of brain disorders based on electroencephalogram (EEG) signals will be provided. The signal processing and machine learning based developed methods will be explained for automated diagnosis of brain disorders. These methods are based on the EEG signals. The comparison of these methods with the existing methods in the literature will be provided. Some future directions related to these presented methods will be given.

Bio Sketch: Dr. Ram Bilas Pachori has been working as a Professor since 2017 at IIT Indore. He was a Visiting Professor at School of Medicine, Faculty of Health and Medical Sciences, Taylorˊs University, Subang Jaya, Malaysia during 2018- 2019. He worked as a Visiting Scholar at Intelligent Systems Research Center, Ulster University, Northern Ireland, UK during December 2014. He is an Associate Editor of Electronics Letters, Biomedical Signal Processing and Control journal and an Editor of IETE Technical Review journal. He is a senior member of IEEE (USA) and a Fellow of IETE (India) and IET (UK). He has more than 200 publications which include journal papers (117), conference papers (65), books (04), and book chapters (16). His publications have around 6500 citations, h index of 43, and i10 index of 103 (Google Scholar, August 2020). He has been listed in the top h index scientists in the area of Computer Science and Electronics by Guide2Research website. His area of research is Signal Processing and Machine Learning.

Tolga Kaya

Title: Validation Study of a Dry Hand Cleaner as a Potential Hand Sanitizer

Abstract: One of the biggest challenges we faced during the pandemic was the lack of hand sanitizing supplies. Forced with looking out for alternative solutions, we conducted a feasibility study on how effective an industrial dirt-cleaning foam could reduce the percentage of germs on hands. Using a glow-in-the-dark hand lotion to represent germs, we took hand images and analyzed these images using MATLAB. Results showed that the industrial cleaning foam could effectively be used to reduce the risk of germ spread.

Bio Sketch: Dr. Tolga Kaya is a Professor and Director of Engineering at the Sacred Heart University. Before joining to SHU, Dr. Kaya was an Associate Professor at Central Michigan University’s School of Engineering and Technology between 2010-2017. Dr. Kaya worked as a research scientist at Yale University from 2007 to 2010 at the School of Engineering and Applied Sciences. He has worked in industry as a technology consultant (Brightwell Corp.) and a senior analog circuit designer (Microelectronics, Inc.) for several years.

Dr. Kaya has a diverse research background and accomplishments on Electrical, Computer, and Biomedical Engineering. His research primarily focuses on wearable devices to monitor physiology of athletes. He is well known in the field of hydration research. Dr. Kaya also focuses on Internet of Things (IoT) devices and 3D printing technologies.

In the past few years, Dr. Kaya also conducted research on education technology and teaching pedagogies. He implements active learning strategies in his classrooms and focuses on student-centered learning. All his courses have significant hands-on components with plenty activities to help students link theory with practice.