Disruptions in Computer Aided Diagnosis: A Deep Learning Perspective

This special session is organised by Upgrad Education (https://www.upgrad.com/gb)

Session Chair:

  • Dr. Rupal Bhargava, upGrad Education Pvt. Ltd, India 

Session Co-Chair:

  • Dr. Rik Das, ACM Distinguished Speaker, India 
  • Assoc. Prof. Dr. Muhammad Ehsan Rana, Asia Pacific University of Technology and Innovation, India  


Recent advancements in Computer Aided Diagnosis (CAD) has leveraged the domain as an active area of research that is pivotal in identifying lethal ailments at its inception. It has achieved remarkable progress with the popularity of recent disruptive trends in machine learning applications. A high level of precision is observed with automated CAD-based systems compared to manual detection of malignancy for terminal diseases challenging the medical science for an extensive period. This has prevented the premature death of many patients due to late detection and several procedural formalities. Therefore, it is pertinent to design efficient algorithms for proposing CAD systems to mitigate the challenges of critical illnesses at an early stage. Researchers are facing multiple challenges in preparing an automated detection system due to lack of training data, sample annotation, region of interest identification, proper segmentation, etc. Fortunately, recent advancements in computer vision and content-based image classification have paved the way for assorted techniques to address the aforementioned challenges and have helped attain novel paradigms for designing CAD systems. Popular deep learning and machine learning application have profusely added in augmenting the detection accuracy. The special session is an attempt to collate novel techniques and methodologies in the domain of content-based image classification and deep learning/machine learning techniques to design efficient computer-aided diagnosis architecture. In this age of seamless connectivity, medical devices are often connected to hospital networks, mobile phones, the internet, etc. Hence it is essential to ensure cybersecurity to prevent patient data as well as the privacy of any individual. The healthcare IoT has an ongoing and augmenting impact in the medical industry by removing the requirement of repeated office visits with telemedicine and related advancements in technology. The ledger technology implemented in blockchain results in the secure transfer of patient medical records, managing the medicine supply chain and helping healthcare researchers unlock genetic code. Big Data Analytics is pivotal in the evolution of healthcare practices and research and is efficiently applied towards aiding the process of care delivery and disease exploration. Cloud computing enables all Big Data operations through the provision of large storage and processing power. Finally, innovations and progress in software development to monitor, analyse and interpret a patient’s medical state is opening a new dimension in healthcare management. Therefore, this special session is aimed to highlight new challenges and probable innovative solutions in the domain of computer-aided diagnosis and healthcare advancements leveraged by IoT, blockchain Big Data and software development. It will also explore the advancements in the domain of cybersecurity to preserve and ensure the privacy of medical records and individual identity as a matter of fundamental rights.


This special session seeks to include original, high-quality contributions to AI systems in healthcare. The key areas of concern include but are not limited to:

  • Malignancy detection in histopathological images using CAD system
  • Identifying region of interest for CAD systems with image segmentation
  • Semantic segmentation for disease detection and malignancy identification
  • Clustering Techniques for designing CAD systems
  • Applications of GANs in computer aided diagnosis
  • Anomaly detection in medical images for computer aided diagnosis
  • Designing CAD systems using Representation Learning Applications with Deep Neural Networks
  • Designing CAD systems applying Transfer Learning Applications with Deep Neural Networks
  • Federated Machine Learning for Computer Aided Diagnosis
  • Efficient techniques for descriptor detection / feature extraction from medical images
  • Attention based deep learning techniques for identifying abnormalities in medical data
  • Computer aided diagnosis for sustainable world
  • Detection and addressing of tampering of interconnected medical devices
  • Addressing ransomware attack on patient data and back office systems
  • Addressing medical insurance fraud by selling of patient data by cyber criminals
  • Innovations in blockchain for clinical trials
  • Addressing challenges with blockchain for medical data privacy compliance
  • Innovations in secured sharing of medical records with blockchains
  • Innovations in IoT based health monitoring
  • Innovations in IoT enabled biosensors
  • Innovations in wearable IoT sensors for healthcare
  • Innovations in IoT in genomics
  • Innovation in IoT devices in pharmaceuticals plants
  • Progress in IoT enabled medical robotics
  • Innovations in software development for healthcare monitoring
  • Managing big data in healthcare
  • Storage optimization for big data in health care
  • Cloud based healthcare data management
  • Innovations in Blockchain for Fraud Detection
  • Privacy-Preserving Blockchain Driven Federated Learning
  • Blockchain Based Solutions to address IoT Middleware Issues
  • Enhancing Blockchain Performance on Cloud

Paper Submission:

Prospective authors are invited to submit full-length papers (not exceeding 6 pages) conform to the IEEE format . All papers will be handled and processed electronically via the EDAS online submission system.