Keynote 1: Web-based intelligent agents for suicide monitoring and early warning
Prof. Zhisheng Huang, Vrije University of Amsterdam, Netherlands
Abstract
Teenage suicide has become one of the important issues of general concern in society. Many young people express various suicidal emotions and wishes through social media (such as Weibo), which provides the possibility to use artificial intelligence to analyze the message on social media to detect the persons who have high risk of suicide. The intelligent agents we developed uses knowledge graph technology to monitor Weibo every day and issue suicide monitoring notifications. The "Tree Hole Rescue Team" organized by us took online suicide rescue operations based on surveillance reports. From July 2018 to June 2021, we have prevented more than 4,765 suicides within two years. This talk will introduce the basic technology of our intelligent "Tree hole" agents and explain how this intelligent agents can be used to detect potential suicides and issue corresponding surveillance notifications.
Keynote 2: The analyses of histopathological images of breast cancers using supervised and unsupervised deep learning techniques
Prof. Juanying Xie, Shaanxi Normal University, China
Abstract
This talk will focus on analyzing the histopathological images of breast cancer using deep learning techniques. It is known that breast cancer is associated with the highest morbidity rates for cancer diagnoses in the world and has become a major public health issue. Early diagnosis can increase the chance of successful treatment and survival. However, it is a very challenging and time-consuming task due to its relying on the experience of pathologists. The automatic diagnosis of breast cancer by analyzing histopathological images plays a significant role for patients and their prognosis. However, traditional feature extraction methods can only extract some low-level features of images, and prior knowledge is necessary to select useful features, which can be greatly affected by humans. Deep learning techniques can extract high-level abstract features from images automatically. Therefore, we introduce it to analyze histopathological images of breast cancer via supervised and unsupervised deep convolutional neural networks respectively in our work. First, we adapted Inception_V3 and Inception_ResNet_V2 architectures to the binary and multi-class issues of breast cancer histopathological image classification by utilizing transfer learning techniques. Then, to overcome the influence from the imbalanced histopathological images in subclasses, we balanced the subclasses using Ductal Carcinoma as the baseline by turning images up and down, right and left, and rotating them counterclockwise by 90 and 180 degrees. The extensive experimental results of the supervised histopathological image classification of breast cancer and the comparison to the results from other studies demonstrate that Inception_V3 and Inception_ResNet_V2 based histopathological image classification of breast cancer is superior to the existing methods. Furthermore, these findings show that Inception_ResNet_V2 network is the best deep learning architecture so far for diagnosing breast cancers by analyzing histopathological images. Therefore, we used Inception_ResNet_V2 to extract features from breast cancer histopathological images to perform unsupervised analysis of the images. We constructed a new autoencoder network to transform the features extracted by Inception_ResNet_V2 to a low dimensional space, so as to extract much higher level abstract features to do clustering analysis for the images. The experimental results demonstrate that our proposed autoencoder network can result in better clustering results than those based on features extracted by Inception_ResNet_V2 network only. This study demonstrates that Inception_ResNet_V2 network based deep transfer learning provides a new way of performing analysis to histopathological images of breast cancer.
Speaker Bio
Juanying Xie is a professor at school of computer science of Shaanxi Normal University, Xi’an, PR China. Her research interests include machine learning, data mining and biomedical data analysis. She has published two monographs, and about 80 journal or conference papers. Her work is highly cited, including one article in the top 1% of ESI and two articles as F5ooo highly cited papers. Furthermore, one of her published articles has become the top 3 hotspot article of "SCIENTIA SINICA Informationis" in 2018. She has been leading several NSFC projects.
Professor Juanying Xie is an associate editor of "Health Information Science and Systems", and one of the editorial board member of "Natural Science edition of the journal of Shaanxi Normal University". In addition, she is a senior member of CCF (China Computer Federation), and a member of CAAI (Chinese Association for Artificial Intelligence). She is the committee member of many Specialties, such as artificial intelligence and pattern recognition of CCF, machine learning of CAAI, bioinformatics and artificial life of CAAI, knowledge engineering and distributed intelligence of CAAI, and image and graphics engineering society of Shaanxi province. She is also the standing committee member of intelligent service of CAAI.
Keynote 3: Addressing the gap between the needs of older people and adoption of technology
Prof. Jeffrey Soar, University of Southern Queensland, Australia
Abstract
Older people have the greatest needs that technology could assist with and there have been expectations of an exciting world of greater independence through technology. Whilst there are seniors who are enthusiastic and sophisticated users of technologies, for many the benefits of digitally-enabled living has largely bypassed them. This presentation will explore the needs of older people, the availability of technologies, challenges to adoption, opportunities for innovation and finding the business case.
Speaker Bio
Prof Jeffrey Soar, is Chair in Human-Centred Technology at USQ. He has 40+ years in ICT; in industry he held CIO positions for national and state government in Australia and New Zealand. In academia he has been research centre director, head of school and school research coordinator, also serving on Academic Board and the Research Committee. His research has focused on smart homes and AI-enabled care. His research achievements include more than 200 publications, 30 grants, seven ARC-funded projects, four Advance Qld/Smart State grants, and a DSTG grant. He teaches project management, business analysis and Agile Methods.
Keynote 4: The integration of clinical pharmacy services with Electronic Medical Records
Dr. Samanta Lalic (Wood), Assistant Deputy Director of Pharmacy - Monash Health Adjunct Research Fellow, Centre for Medicine Use and Safety, Faculty of Pharmacy and Pharmaceutical Sciences, Monash University
Abstract
The adoption of Electronic Medical Records (EMR) within pharmacy departments in hospitals has led to improvements in medication safety. The integration of EMR hasn't been met without its challenges. Achieving and utilising the full potential of EMR required pharmacists and other healthcare providers to actively engage with the system but also be actively involved in the creation of a system that was fit for purpose. A perspective of profession-specific barriers and benefits to the use of EMR within a large hospital network will be discussed.
Speaker Bio
Dr Lalic/Wood [BPharm (Hons), MPharmPrac, PhD] is a senior clinician at Monash Health and an early career clinician researcher at both Monash Health and the Centre for Medicine Use and Safety, Monash University. Currently, Dr Lalic leads the research and education programs at the pharmacy department at Monash Health. Prior to her role as the Assistant Deputy Director of Pharmacy at Monash Health, Dr Lalic worked across various specialities including general medicine and geriatrics at another large tertiary hospital network. In addition to clinical expertise, Dr Lalic has expertise in research methodology and specifically analysing PBS data and electronic medical data. Together with Professor Bell and Dr Ilomaki, Dr Lalic was the first to apply group-based trajectory modelling as a novel technique to investigate persistence to opioids in Australia. Through her PhD, Dr Lalic developed expertise in analysing administrative datasets. Dr Lalic has published her work in high impact journals including Pain, Addiction, JAMDA and JACI: In practice. Dr Lalic has secured over 650k in research funding.
