Machine Learning in Small Data
for Human and Process Systems



Leveraging small data to create value that cannot be obtained from big data

The third AI boom has progressed around deep learning, based on the naive belief that if big data, which can be collected inexpensively, are processed with a high-speed computer, it would be possible to create a machine that is smarter than humans. On the other hand, human experience, implicit knowledge possessed only by a small number of experts, and data whose occurrence itself is rare or whose collection is high-cost, are being forgotten.

We are intentionally focusing on small data and are carrying out analyses that proactively incorporate human experience and implicit knowledge, with the aim to crate new value that cannot be obtained from big data. Specifically, we use small data analyses to develop novel medical devices and realize innovative production systems.


We develop algorithms that utilize data collection and machine learning, using small data as a key concept. We aim to create the best research environment, such as always providing students with desktop PCs / MacBooks with the latest specifications, providing s cloud computing environment, and facilitating participation in international

Research Topics


Sleep Disorder

The public is greatly concerned with sleep, exemplified by NHK introducing the term sleep debt, but not many people are aware of their own sleep disorders. We are developing an algorithm that is capable of easily screening sleep disorders by analyzing sleep polygraph data.


Epilepsy is a disease that causes ictal events such as convulsions and loss of consciousness, wherein accidents and injuries caused by the events become a problem. Based on clinically collected data, we are developing systems that predict epileptic seizures as well as AI that supports epilepsy care by non-specialists.

Medical Big Data Analysis

Various medical data, such as medical checkups, routine medical examinations, and patient registries are accumulated on a daily basis, but they are not currently being effectively utilized. In our research, we explore the causes of diseases by digging up and analyzing these underutilized medical data.


With the proliferation of smartphones and miniaturization of sensors, it has become possible to easily measure biological signals, such as heart rate and skin resistance, in daily life. We are developing mechanisms that change the user experience (UX) in entertainment, such as video games, by utilizing such biometric technologies.

Fostering AI Personnel

Application of AI in the medical field has been called for for quite a while, but incorporation of AI in medical practice is lagging. The cause for this is the fact that there are no personnel with detailed knowledge of machine learning on the medical side, as well as the fact that understanding of medicine is lacking on the engineer side. We aim to foster personnel that are able to connect medical personnel and engineers.

Soft Sensors

In a production process, there are many variables whose measurement is high-cost, but online measurement of such variables is required in order to ensure quality control and safety. We are studying the methodology of soft sensors that estimate variables that are difficult to measure, from variables for which online measurement is easy.


Associate Professor




MAIL Nagoya University, Graduate School of Engineering, Department of Materials Science and Engineering
Furocho, Chikusa-ku, Nagoya-shi, AICHI 464-8601

Nagoya University, Higashiyama Campus, Faculty of Engineering, Building 1, Suite 401

Click here for the campus map (Building: B2-5)


Please use the form below to inquire about research content, and to request a laboratory visit or consultation on admissions