Students from Sepuluh Nopember Institute of Technology (ITS) in Indonesia have developed devices likely capable of detecting disorders such as anaemia and neurological disorders.
In one project, a team of students came up with a sensor-based AI-powered detector that can measure haemoglobin levels to predict the likelihood of anaemia in patients with Systemic Lupus Erythematosus. Called Hemoglobest, the non-invasive haemoglobin test uses five spectrums of light, which is “more effective compared to oximeters” that use only two light spectrums, one student was cited as claiming.
Another project is a rapid diagnostic microfluidic biosensor tool that is possibly capable of detecting neurological disorders. The detector called NeuroCube uses colourimetry, acting like a litmus test that reacts to neurotransmitter compounds in urine samples. The colour change indicates the concentration level of such compounds as dopamine, glutamate, and Nicotinamide Adenosine Dinucleotide Hydrogen to detect neurological disorders, including dementia, OCD, ADHD, bipolar disorder, schizophrenia and Alzheimer’s disease.
A team of researchers from Singapore and Kazakhstan have combined AI and heat imaging technologies to assist in potentially detecting breast cancer early.
They developed a computer program called Physics-informed Neural Network that uses AI to analyse heat patterns in thermal infrared breast images and flag suspicious findings indicative of malignant tumours “within five minutes.” It was trained and tested using thousands of infrared breast scans of patients with or without harmful tumours in Kazakhstan and achieved a detection accuracy of 91%.
“Tumours, including breast cancer tumours, often have distinct metabolic activity and blood supply compared to normal tissue. As a result, they may generate more heat or have different thermal properties,” Dr Anna Midlenko, a clinical instructor in the Department of Surgery at Nazarbayev University, explained their study, whose findings have been published in Computer Methods and Programs in Biomedicine journal.
Eddie Ng Yin Kwee, associate professor from the School of Mechanical and Aerospace Engineering at Nanyang Technological University Singapore, said their team is now researching to further enhance the AI program to predict tissue properties and tumour sizes and locations via inverse-bioheat transfer techniques, hoping that it can serve as a “new portable AI tool for the early detection of breast cancer and breast self-examination.”
A team of researchers from the Korea Advanced Institute of Science and Technology claims to have created a sweat-resistant electromyography (EMG) sensor which allows for long-term stable control of wearable robots for rehabilitation.
Existing sensors for wearable robotic rehabilitation systems often deteriorate in signal quality over time and are easily affected by the wearer’s skin conditions. The KAIST team attempted to address such a limitation by creating a stretchable and adhesive microneedle sensor which can “sense physiological signals at a high level without being affected by the state of the user’s skin.”
The sensor features hard microneedles that penetrate through the stratum corneum, which has high electrical resistance and thus has lower contact resistance with the skin and can obtain high-quality electrophysiological signals regardless of contamination.
“Through this, we will be able to control wearable robots with higher precision and stability, which will help the rehabilitation of patients who use robots,” explained KAIST professor Jae-Woong Jung, who led the research. Their findings were published in the journal, Science Advances.