Research

Medical and Health Care Applications Aided by IT and AI

Information technology (IT) and Artificial Intelligence(AI) have become the primary technology in the current medical practice and health care by using high performance computers with low cost. In it, Especially, Soft computing defined by Prof. Lotfi A Zadeh (UC Berkeley) is a consortium of methodologies that exploit a tolerance for imprecision, uncertainty and partial truth. Results are achieved with tractability, robustness and a rapport with reality. These results are often better than that what is achieved through the exclusive use of conventional (hard) computing. The principal members of soft computing are for the most part synergistic and complementary. The main ideas of soft computing were brought forth by the confluence of fuzzy logic, neural networks, and probabilistic reasoning. Other ingredients include evolutionary algorithms, rough sets, chaos theory, belief networks and, although only partially, learning theory. In soft computing, each constituent contributes a distinct methodology for addressing problems in its domain. This is done in a complementary, rather than competitive way. Especially, I am now interested in realization of human reasoning, multiple-valued logic design and medical imaging applications.

I have been studying medical and health related system since 1995. My current research topics are:

MEDICAL SYSTEM: Medical imaging has evolved at an explosive rate in the past few years. High-resolution, three-dimensional anatomical information can now be obtained in a routine manner with magnetic resonance imaging(MRI) and computer-aided tomography(CT) and Ultrasonic Imaging. Three-dimensional functional imaging of blood flow and metabolic information can be obtained from positron emission tomography(PET) images, and functional four-dimensional electrophysiology (EEG) and evoked potential information can be obtained using high-speed computers. Especially, AI provides wonderful results for static image analysis. In my labo., dynamic (3D) image analysis is studying.

HEALTH CARE SYSTEM : Due to the huge cost of medical fee of about 40 % in national budget, Japan faces to save the cost. The most promising way to save the cost is keeping the all Japanese people healthy. To control health condition, a wearable device including smart phones receives much considerable attentions. Especially, smartphone software aimed at weight control, sleeping recording, food calorie calculation, pedometer, blood pressure recording is used for many users. Wearable devices to record life log are also popular. While, all most Japanese people receives medical checkup. Medical checkup over 40 years old people employs breast X-ray, weight and height, blood pressure, eye test, blood test, and stomach X-ray. These data would be useful to find how to keep them healthy. Cloud computing gives us a wonderful solution to solve this problem. Health related system and its market with information technology would be receive much more attentions in the world. This direction is the most promising way to keep all people healthy in the world.

(1) Infertility Study:

Infertility is defined clinically in women and men who cannot achieve pregnancy after 1 year of having intercourse without using birth control, and in women who have two or more failed pregnancies. Studies suggest that after 1 year of having unprotected sex, 15% of couples are unable to conceive, and after 2 years, 10% of couples still have not had a successful pregnancy. In couples younger than age 30 who are generally healthy, 20% to 37% are able to conceive in the first 3 months. In this talk, I will focus on the fusion research of computer science and reproductive medicine, and discuss the usefulness. I will talk about the following topics.

(1) Detection of thick seminiferous tubules in the testicles: In micro-TESE surgery is performed using a microscope, and the seminiferous tubules that produce sperm are removed, and the sperm is collected from the removed seminiferous tubules. This sperm collection surgery for patients with non-obstructive azoospermia requires incision of the testes. This damage needs to be minimized, and this cost must be saved. Therefore, it is required for both surgeons and patients to non-invasively detect the presence of sperm in the testes before this operation. In this study, we detect the seminiferous tubules with a diameter of 250-300μm that can recover sperm using ultrasound technique. However, currently used ultrasonic waves of about 8 MHz cannot find a fine tube of this diameter by the low resolution. Thus, we found a characteristic that the peak frequency of the ultrasonic reflected wave is proportional to the reciprocal of the diameter, and estimated the diameter of the seminiferous tubule present in the testicle based on fuzzy inference.

(2) Supporting of ovum collection surgery: At the time of surgery to collect ova, an ultrasonic system is generally used. However, the follicles watched by the ultrasound system do not always contain ovum, and there are follicles that do not contain ovum is called vacuole. It is not possible in terms of resolution to confirm the vacuoles with ultrasound images before their removal, and the presence or absence of ovum is only known after collection. Therefore, we developed detection software of vacuoles using AI.

(3) Determining the insertion position of sperm that does not cause rupture on the ovum: In micro-insemination, a sperm is directly injected into an ovum collected from a woman using a pipette. When injecting sperm, the oval cell membrane is usually sufficiently extended with a pipette, and then rupture is performed by piezo pulse. However, rupture may occur during the development of the membrane, in which case the insemination rate is reduced. Therefore, we quantitatively evaluate the effect of puncture position on ruptured membrane from ovum images during micro-insemination, and develop a system that can perform puncture based on the evaluation.

(4) Analysis of peristalsis of the uterus: The uterus performs a movement called uterine peristalsis to transport sperm. It is known that the direction and frequency of uterine peristalsis change according to the menstrual cycle. We developed a system to evaluate peristalsis of the uterus by Cine-MRI analysis, and an evaluation system of peristaltic movement frequency by the clinical ultrasonic image analysis.


(2) Motion Analysis Study


(3) Swallowing Dynamics Study


Key words:

1. Medical Imaging System

2. Ultrasonic Diagnostic Imaging

3. Health Monitoring System

4. Medical Informatics

5. Medical Signal Processing


BOOKS

  1. Computational Intelligence Volume II (H. Ishibuchi ed.) Computational Intelligence and Medical Applications, Encyclopedia of Life Support Systems , UNESCO EOLSS, pp.172-201, Dec. 2015

  2. Fuzzy Logic Theory and Applications: Part Ⅰ and Part Ⅱ (Lofti A Zadeh and Rafik A Aliev ed.), Fuzzy logic in medicine, World Scientific, pp. 531-546, Dec. 2018. ISBN: 978-981-3238-17-6 (hardcover) , ISBN: 978-981-3238-19-0 (ebook)

  1. Systems of Systems Engineering: Principles and Applications in Computational Intelligence ( Mo Jamshidi ed.), Medical and health management system of systems, CRC press, pp.233-250, 2008.