CONFERENCE RESEARCH
INGENIEUR
CONFERENCE RESEARCH
INGENIEUR
Application of Artificial Intelligence in Healthcare
By Ir . Dr Goh Choon Hian Ir . Dr Chee Pei Song Department of Mechatronics and BioMedical Engineering Lee Kong Chian Faculty of Engineering and Science Universiti Tunku Abdul Rahman Malaysia
Artificial intelligence ( AI ) is revolutionising the healthcare industry by providing innovative ways to understand , diagnose , and treat diseases . AI algorithms are being applied to various domains in the healthcare sector , from drug discovery to clinical trials , clinical research , medical diagnosis , and medical treatment . In addition , with the increasing amount of data generated daily , AI is becoming a crucial tool in handling this data and making sense of it . The goal of AI in healthcare is to improve patient outcomes , advance medical research , and reduce the cost of healthcare delivery .
Artificial Intelligence ( AI ) has been making significant contributions to the biomedical field in recent years . AI techniques such as machine learning , deep learning , and natural language processing are used to analyse and interpret biomedical data . These techniques allow for the creation of models that can predict outcomes , identify patterns , and make decisions based on enormous amounts of data .
Medical Imaging
Medical imaging is one of the areas where AI is having a significant impact . AI algorithms can be used to analyse medical images , such as X-rays , Magnetic Resonance Imaging ( MRI ) scans , and Computerised Tomography ( CT ) scans to detect diseases and abnormalities . This can lead to improved accuracy and efficiency in diagnosis and can also help to reduce the workload of radiologists .
In the field of genomics , AI is being used to analyse enormous amounts of genetic data to identify disease-causing mutations and develop personalised treatments . The COVID-19 pandemic has presented significant obstacles for healthcare institutions worldwide , leading to the need for effective strategies . Thoracic imaging has emerged as a valuable tool in the diagnosis , prediction , and treatment of COVID-19 patients with moderate to severe symptoms or deteriorating respiratory conditions ( Liu et al ., 2022 ). In a review article authored by Liu et al ., the use of machine learning and deep learning models was examined , revealing promising overall performance based on research findings .
In Universiti Tunku Abdul Rahman ( UTAR ), there is a group of researchers working on breast cancer diagnosis . The use of computerassisted deep learning in Invasive Ductal Carcinoma ( IDC ) grading classification systems has demonstrated that deep learning can attain dependable accuracy in IDC grade classification using histopathology images . However , there is limited research that compares the performance of different Convolutional Neural Network ( CNN ) designs in IDC classification . In response , a study was conducted to evaluate seven selected CNN models : EfficientNetB0 , EfficientNetV2B0 , EfficientNetV2B0-21k , ResNetV1-50 , ResNetV2-50 , MobileNetV1 , and MobileNetV2 using transfer learning . The study utilised the corresponding feature vector from TensorFlow Hub for each pre-trained CNN architecture , incorporating it with dropout and dense layers to form a full CNN model . The results showed that the EfficientNetV2B0-21k model
52 VOL 95 JULY-SEPTEMBER 2023