Applications of AI in medicine: Intelligent Health AI 2020’s Impressions
Before 2020 every September the world’s brightest AI health brains from pharmaceutical, biotech, medtech, health provisions, clinicians, tech companies, startups, investment and science swarmed to Basel, Switzerland to participate in the only large-scale, global summit focused purely on applications of AI in medicine – Intelligent Health AI. Because of the pandemic, this year the event took place entirely online. On the hopin platform. Besides, with attendees from 164 countries, participation of 3k+ world leaders, 200 speakers, 4 cyberspace innovators & startups presenting. With the involvement of 48 investors, and coverage by 38 international media and TV.
Intelligent Health AI – AI in medicine event
Generally, the least the event was huge and packed with dynamic and interactive content. Most important is that there were heated discussions, tech demos and interactive workshops streamed from all corners of the world leaving little time to catch breath between sessions. Additionally, the organizers went to great lengths to provide space for the participants to connect the AI and medicine communities. For example with the networking roulette function where one could introduce oneself to other online delegates in 3-minute speed-dating style sessions. And also video and text chat functionalities plus interactive workshop sessions to leverage the outstanding community connections and take away some tangible ROI from the world’s leading AI in medicine event. In the Cyberspace area one could find virtual booths hosted by Intelligent Health AI sponsors and partners to gather information and interact with the hosts.
We are looking forward to next year’s edition in September 2021.
Intelligent Health AI 2020 – themes
Key themes for Intelligent Health AI 2020 included:
- The future of healthcare – where can AI take us with practical, real-world AI use cases
- Data everywhere – how to get the data we need to drive forward AI applications?
- COVID-19 – What has the last 9 months taught us about leveraging digital technologies?
- Regulators perspective – FDA and responding to pandemics
- Deep dive AI tech talks and demos – fundamental breakthroughs using AI which will change lives and reshape healthcare
- AI health transformers – hear from the CEOs and the Founders of the most disruptive AI health tech companies of tomorrow
- Healthcare without borders – democratising access and AI’s role in combating inequalities in health
- Creating an AI enabling landscape – applying AI to achieve whole person care
- The role of AI in genomics – data and humans in perfect harmony
Applications of AI in medicine: our impressions
This year our team was represented by our Machine Learning software engineers Bartosz Machura, Krzysztof Kotowski and Wojciech Malara and here are some of their impressions.
Applications of AI in medicine – quotes
Bartosz Machura – unexpected applications of AI in medicine
![Applications of AI in medicine - the quote text on applications of AI in medicine by Graylight Imaging's expert](https://graylight-imaging.com/wp-content/uploads/2024/04/Applications-of-AI-in-medicine-the-quote.webp)
Day 1: Brilliant conference! Many interesting topics were covered that showed me some unexpected applications of AI in medicine, such as: virtual staining of histopathological samples, attempts of COVID-19 detection from cough or real-time analysis of agents released from an incision during the surgery. Especially worth mentioning were three speeches:
- Digital surgery – How surgeons will use data to inform intraoperative decision making, – Dr. James Kinross, Senior Lecturer and consultant in Colorectal Surgery, Department of Surgery and Cancer, St. Mary’s Hospital, Imperial College London
- How digital technologies can be leveraged for the increasing threads of future pandemics – Dr. Sonny Kohli, Internal Med/Critical Care Physician, Oakville Hospital, Assistant Prof (adj.), McMaster University, Founding Faculty, Singularity University Canada
- Transforming skin cancer diagnosis with deep neural networks by Andre Esteva, PhD, Head of Medical Al, Salesforce Research
Krzysztof Kotowski – Artificial General Intelligence may be the next game changer
![Applications of AI in medicine - quote from the conference text on artificial intelligence and artificial general intelligence by Graylight Imaging ML expert](https://graylight-imaging.com/wp-content/uploads/2024/04/Applications-of-AI-in-medicine-quote-from-the-conference.webp)
After two days of this conference, I am even more excited to be the part of this community. COVID-19 pandemic forced us to meet online, but also emphasized the real need for robust AI supporting healthcare systems in these hard times. The number of innovative AI applications in Healthcare surprised me – from automated medical documentation, through medical image segmentation, to autonomous surgical robots. The problem is that still only few make a real impact on the global healthcare system. Artificial General Intelligence (great talk from Gary Marcus) may be the next game changer.
One of the strong aspects of the conference was the focus on regulatory requirements of AI solutions (Andrea Biasiucci from Confinis – thanks for the insights into MDCG Guidance 2020-1 on Clinical Evaluation of Medical Device Software). The options of visiting virtual booths and networking gave many opportunities to meet experts from all over the world.
Wojciech Malara – limitations of pure deep learning approach
![Applications of AI in medicne text on applications of artificial intelligence and covid by Graylight Imaging expert](https://graylight-imaging.com/wp-content/uploads/2024/04/Applications-of-AI-in-medicne.webp)
Summary
To summarize, the talk that was particularly interesting to me was “COVID-19 Should be a Wake-up Call for AI” by Gary Marcus. It was based on an article “The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence“. In the talk, prof. Marcus highlighted some limitations of the pure deep learning approach. This is great for learning complex statistical correlations and perceptual classification but performs poorly in terms of reasoning and language understanding. In general, deep learning cannot extrapolate beyond training data. As a step towards more robust AI the author proposed “a hybrid, knowledge-driven, reasoning-based approach, centered around cognitive models”.
Contact us if you have any questions!