1) Please tell us a little bit about yourself
I am a general dentist and a PhD student, currently also working as a Clinical Insights Analyst for a start-up data analytics company operating in the health care industry. Outside of work and study – I enjoy hiking, bouldering, road trips and theatre.
My PhD research involves exploring the use of natural language processing (NLP) methods on clinical notes in dentistry for analysis, decision support systems and population health surveillance. I am interested in improving the quality of clinical notes and how information in clinical notes can be made more useful and serve as real world data.
2) How do you define digital health?
Digital health is a very broad term that encompasses computer vision, robotics, wearables, machine learning and natural language processing. I think of digital health essentially as a tool to improve clinical workflows, patient outcomes and health care delivery.
3) What do you think will enable digital health projects and innovations to succeed?
Collaboration amongst professionals from diverse backgrounds is critical to making widespread improvements in health care. Learning from each other’s expertise and developing co-designed solutions has the greatest probability for success and changing the health care paradigm.
In my experience, the most successful projects are those where diverse perspectives are incorporated into the product. I really enjoy working in teams involving data scientists, data engineers, clinicians from different disciplines and incorporating the patient’s voice – we learn so much from each others’ experiences.
4) What do you think are the biggest challenges facing digital health at the moment?
The lack of integration of health data is challenging from an analysis and research perspective. – not only integration of health records, but also user-generated data. This is slowly improving, but by and large health professionals are still working in siloes. For example, a shared electronic record between a patient’s GP and dentist would mean patients don’t have to repeat their medical history to their dentist – it’s well documented that patient-reported histories can be quite inaccurate compared with their medical record.
With respect to machine learning (ML) applications, I think there are several challenges to successful clinical implementation including identifying solutions that have a positive impact on patient care, integrate seamlessly with clinical workflows or improve workflows, black box models and liability, evaluating ML on clinically relevant performance metrics, and long-term generalisability. It’s great to see that recently published work addresses a lot of these key areas.
5) Do you have any interesting resources or helpful networks people should know about?
Can’t go past DHIN as an incredibly informative resource, and for connecting with others working in the digital health space.
Leave A Comment