Source: Okayama University (JAPAN), Public Relations Division
For immediate release: 28 March 2021
Okayama University research: Studying Parkinson’s disease with face-recognition software

(Okayama, 28 March) Researchers at Okayama University report in Brain Supplement that artificial-intelligence technology can detect facial characteristics of Parkinson’s disease. The faces of patients were systematically found to look older and expressionless.

Parkinson’s disease is a brain disorder leading to motor symptoms including shaking, stiffness and difficulty with walking, as well as mental symptoms such as depression, memory problems and fatigue. Usually, the syndrome also includes facial abnormalities known as ‘facial masking’ — an affected person’s face has a mask-like expression. Given the recent progress in face-recognition tools based on artificial intelligence (AI), Professor ABE Koji and colleagues from Okayama University explored whether AI technology can be used to detect facial changes in patients with Parkinson’s disease.

The researchers worked with 96 healthy (control) subjects and 97 patients with Parkinson’s disease. The face of each participant was photographed and then analyzed with AI software. For each facial photograph, the program produced a set of attributes such as age, gender and emotion.

By looking at the ‘age gap’, defined as the appearance age (as determined by the AI software) minus the real age, the scientists found that the appearance of patients with Parkinson’s disease made them look older by an average of 2.4 years. For male patients, the average age gap was even 3.4 years. Another observation was that elder patients tended to have a smaller age gap than younger patients.

Regarding emotions, Abe and colleagues found that for the patients with Parkinson’s disease, expressionless faces were significantly more frequent than for the healthy control subjects (89% vs. 77%, respectively), and that happy faces were significantly less frequent (5% vs. 19%, respectively). Other emotions, such as contempt, surprise, disgust, anger and fear were not found to differ between the two groups.

The condition of the participants’ facial skin was also analyzed based on photographs, with the aim of taking skin features such as stains, wrinkles and eye shadow into account. No significant differences between the skin of healthy subjects and Parkinson’s disease patients were found, though. The scientists believe that the employed smartphone application did not focus on the oiliness of facial skin.

The overall conclusion of Professor ABE and colleagues is that “Parkinson’s disease patients looked older and expressionless using publically available AI face recognition software”. They point out, however, that the accuracy of facial recognition software depends on gender and skin color, which leads to ethical concerns. Quoting the scientists: “Although face recognition is a remarkable technology, its ethical risk should also be resolved for clinical application.”

Background
Parkinson’s disease : In patients suffering from Parkinson’s disease, the progressive loss of the function or structure of neurons (brain cells) leads to a disorder of the central nervous system, affecting its motor system. Tremor, slowness of movement and difficulties with walking are among the main symptoms in the early stages of Parkinson’s, with dementia being common at more advanced stages.

Another symptom often associated with Parkinson’s disease is the loss of facial expressions, known as hypomimia. It refers to a patient often having a fixed, mask-like expression. Hypomimia is a consequence of the progressive loss of motor control extending to the facial muscles. The condition often estranges acquaintances, and can make it difficult for care partners to interact with the patient, as they cannot always properly assess the latter’s mood.

Now, Professor ABE Koji and colleagues from Okayama University have shown that artificial-intelligence applications can characterize the faces of Parkinson’s disease patients as looking older and expressionless.


Reference
Koh Tadokoro, Toru Yamashita, Yusuke Fukui, Zhihong Bian, Xinran Hu, Mami Takemoto, Ryo Sasaki, Namiko Matsumoto, Emi Nomura, Ryuta Morihara, Yoshio Omote, Nozomi Hishikawa, Koji Abe. Detecting facial characteristics of Parkinson’s disease by novel artificial intelligence (AI) softwares. Brain Supplement, 2021;3:1-7.
Detecting facial characteristics of Parkinson’s disease by novel artificial intelligence (AI) softwares
Reference (Okayama Univ. e-Bulletin): Professor ABE’s team
OU-MRU Vol.74：Rising from the ashes—dead brain cells can be regenerated after traumatic injury
OU-MRU Vol.79：Novel blood-based markers to detect Alzheimer’s disease
OU-MRU Vol.87：Therapeutic potential of stem cells for treating neurodegenerative disease


Correspondence to
Professor ABE Koji, M.D., Ph.D.
Department of Neurology, Graduate School of Medicine, Dentistry and Pharmaceutical
Sciences, Okayama University, 2-5-1 Shikata-cho, Kita-ku, Okayama 700-8558, Japan
E-mail: abekabek(a)cc.okayama-u.ac.jp
For inquiries, please contact us by replacing (a) with the @ mark.
//www.okayama-u.ac.jp/user/med/shinkeinaika/english.html

Further information
Okayama University
1-1-1 Tsushima-naka , Kita-ku , Okayama 700-8530, Japan
Public Relations Division
E-mail: www-adm(a) adm.okayama-u.ac.jp
For inquiries, please contact us by replacing (a) with the @ mark.
Website: //www.okayama-u.ac.jp/index_e.html
Okayama Univ. e-Bulletin: //www.okayama-u.ac.jp/user/kouhou/ebulletin/
We love OKAYAMA UNIVERSITY: https://www.youtube.com/watch?v=7cXlttQIk3E
Okayama University Image Movie（2020）
https://www.youtube.com/watch?v=BKRoF0tffmA
Okayama University supports the Sustainable Development Goals: https://sdgs.okayama-u.ac.jp/en/


Okayama University Medical Research Updates （OU-MRU）
The whole volume : OU-MRU (1- )
Vol.1：Innovative non-invasive ‘liquid biopsy’ method to capture circulating tumor cells from blood samples for genetic testing
Vol.2：Ensuring a cool recovery from cardiac arrest
Vol.3：Organ regeneration research leaps forward
Vol.4：Cardiac mechanosensitive integrator
Vol.5：Cell injections get to the heart of congenital defects
Vol.6：Fourth key molecule identified in bone development
Vol.7：Anticancer virus solution provides an alternative to surgery
Vol.8：Light-responsive dye stimulates sight in genetically blind patients
Vol.9：Diabetes drug helps towards immunity against cancer
Vol.10：Enzyme-inhibitors treat drug-resistant epilepsy
Vol.11：Compound-protein combination shows promise for arthritis treatment
Vol.12：Molecular features of the circadian clock system in fruit flies
Vol.13：Peptide directs artificial tissue growth
Vol.14：Simplified boron compound may treat brain tumours
Vol.15：Metamaterial absorbers for infrared inspection technologies
Vol.16：Epigenetics research traces how crickets restore lost limbs
Vol.17：Cell research shows pathway for suppressing hepatitis B virus
Vol.18：Therapeutic protein targets liver disease
Vol.19：Study links signalling protein to osteoarthritis
Vol.20：Lack of enzyme promotes fatty liver disease in thin patients
Vol.21：Combined gene transduction and light therapy targets gastric cancer
Vol.22：Medical supportive device for hemodialysis catheter puncture
Vol.23：Development of low cost oral inactivated vaccines for dysentery
Vol.24：Sticky molecules to tackle obesity and diabetes
Vol.25：Self-administered aroma foot massage may reduce symptoms of anxiety
Vol.26：Protein for preventing heart failure
Vol.27：Keeping cells in shape to fight sepsis
Vol.28：Viral-based therapy for bone cancer
Vol.29：Photoreactive compound allows protein synthesis control with light
Vol.30：Cancer stem cells’ role in tumor growth revealed
Vol.31：Prevention of RNA virus replication
Vol.32：Enzyme target for slowing bladder cancer invasion
Vol.33：Attacking tumors from the inside
Vol.34：Novel mouse model for studying pancreatic cancer
Vol.35：Potential cause of Lafora disease revealed
Vol.36：Overloading of protein localization triggers cellular defects
Vol.37：Protein dosage compensation mechanism unravelled
Vol.38：Bioengineered tooth restoration in a large mammal
Vol.39：Successful test of retinal prosthesis implanted in rats
Vol.40：Antibodies prolong seizure latency in epileptic mice
Vol.41：Inorganic biomaterials for soft-tissue adhesion
Vol.42：Potential drug for treating chronic pain with few side effects
Vol.43：Potential origin of cancer-associated cells revealed
Vol.44：Protection from plant extracts
Vol.45：Link between biological-clock disturbance and brain dysfunction uncovered
Vol.46：New method for suppressing lung cancer oncogene
Vol.47：Candidate genes for eye misalignment identified
Vol.48：Nanotechnology-based approach to cancer virotherapy
Vol.49：Cell membrane as material for bone formation
Vol.50：Iron removal as a potential cancer therapy
Vol.51：Potential of 3D nanoenvironments for experimental cancer
Vol.52：A protein found on the surface of cells plays an integral role in tumor growth and sustenance
Vol.53：Successful implantation and testing of retinal prosthesis in monkey eyes with retinal degeneration
Vol.54：Measuring ion concentration in solutions for clinical and environmental research
Vol.55：Diabetic kidney disease: new biomarkers improve the prediction of the renal prognosis
Vol.56：New device for assisting accurate hemodialysis catheter placement
Vol.57：Possible link between excess chewing muscle activity and dental disease
Vol.58：Insights into mechanisms governing the resistance to the anti-cancer medication cetuximab
Vol.59：Role of commensal flora in periodontal immune response investigated
Vol.60：Role of commensal microbiota in bone remodeling
Vol.61：Mechanical stress affects normal bone development
Vol.62：3D tissue model offers insights into treating pancreatic cancer
Vol.63：Promising biomarker for vascular disease relapse revealed
Vol.64：Inflammation in the brain enhances the side-effects of hypnotic medication
Vol.65：Game changer: How do bacteria play Tag ?
Vol.66：Is too much protein a bad thing?
Vol.67：Technology to rapidly detect cancer markers for cancer diagnosis
Vol.68：Improving the diagnosis of pancreatic cancer
Vol.69：Early gastric cancer endoscopic diagnosis system using artificial intelligence
Vol.70：Prosthetics for Retinal Stimulation
Vol.71：The nervous system can contribute to breast cancer progression
Vol.72：Synthetic compound provides fast screening for potential drugs
Vol.73：Primary intraocular lymphoma does not always spread to the central nervous system
Vol.74：Rising from the ashes—dead brain cells can be regenerated after traumatic injury
Vol.75：More than just daily supplements — herbal medicines can treat stomach disorders
Vol.76：The molecular pathogenesis of muscular dystrophy-associated cardiomyopathy
Vol.77：Green leafy vegetables contain a compound which can fight cancer cells
Vol.78：Disrupting blood supply to tumors as a new strategy to treat oral cancer
Vol.79：Novel blood-based markers to detect Alzheimer’s disease
Vol.80：A novel 3D cell culture model sheds light on the mechanisms driving fibrosis in pancreatic cancer
Vol.81：Innovative method for determining carcinogenicity of chemicals using iPS cells
Vol.82：Making memories — the workings of a neuron revealed
Vol.83：Skipping a beat — a novel method to study heart attacks
Vol.84：Friend to Foe—When Harmless Bacteria Turn Toxic
Vol.85：Promising imaging method for the early detection of dental caries
Vol.86：Plates and belts — a toolkit to prevent accidental falls during invasive vascular proceduresa
Vol.87：Therapeutic potential of stem cells for treating neurodegenerative disease
Vol.88：Nanotechnology for making cancer drugs more accessible to the brain