Last Tuesday, the Deep Learning in Healthcare Summit London took place at LSO St Luke’s.
RE•WORK hosted 40 speakers and 200 attendees over the course of the 2-day summit to explore the latest developments and applications of Deep Learning (DL) within healthcare, medicine and diagnostics.
Beginning the opening session “Deep Learning & Healthcare in Practice” was Assistant Professor Michael Kuo from UCL with Towards the Development of Clinically Relevant Applications of DL in Healthcare.
Michael emphasized how medicine is an information science that relies on context-specific data. Its practice and the process of medical diagnosis is like a funnel: A flow down of differentials that increases the specificity of the possibilities at each stage. The complex interaction between various factors within the diagnostics search space and the treatment search space make the outcome of a diagnosis or a treatment largely variable. Thus, areas in which he finds ideal to apply DL-based methods are in medical imaging, genomics, clinical data and drug discovery; where context-specific data can be better refined against a database of medical knowledge. Nonetheless, these applications will be riddled with challenges of data quality and quantity, stemming from the limited number of cases per disease diagnosis.
Complementing this line of thought was Professor Neil Lawrence, from the University of Sheffield. He spoke with a focus on the Challenges for Delivering Machine Learning in Health. Referring to the hype cycle, Neil suggests that the healthcare sector is at the peak of inflated expectations, especially with regard to big data applications in healthcare.
Big data is most sharply in focus for personalised health. At the same time, its use is limited. With the current breadth of data, it is possible to quantify, to an increasing degree, the characteristics of individuals but less so to characterise society. More is measured but less is understood on a significant level. Furthermore, data needs to be cleaned before it can be used in machine learning. This time-consuming task improves results, but how valuable or how appropriate the data is, to start with, remains relatively unknown. Finally, issues of privacy, loss of control and the ethics of using patient data, continue to surround the application of Machine Learning in healthcare.
But there remains “a great promise for personalised health” as proven by the magnitude of research and breakthroughs in the sector: Oladimeji Farri, Senior Research Scientist at Philips Research uses DL in diagnostic inferencing and clinical paraphrasing while Polina Mamoshina, Research Scientist at Insilico Medicine, applies DL to biomarker development.
In case you’ve missed the summit, you can watch all of the fascinating presentation videos, view the slides and learn more from exclusive interviews with our speakers on the RE•WORK Video Hub soon. Register here for on-demand access or contact Chloe on email@example.com to receive additional discount for multiple-event access.
The next Deep Learning in Healthcare Summit will take place in Boston on 25-26 May. Early Bird pass are on sale until 31 March. Hear from speakers such as David Plans, CEO of BioBeats, Christhian Potes, Senior Scientist at Philips Research, and Muyinatu Bell, Assistant Professor at John Hopkins University. Register here now to save 400 USD.