Balance Health Uses AI to Reimagine Diabetes Patient Engagement
Serial entrepreneur Pete Lomas built on his own struggle with Type 1 diabetes, and his extensive experience in the CGM market, to develop a platform that uses a unique manual ML knowledge management and acquisition technique to replicate expert diabetes clinicians’ decision making and give context-based guidance to patients with T1D.
Investors, learn how you can back Health Transformers like Pete Lomas.
Challenge
Pete Lomas is a born entrepreneur. When he was diagnosed with diabetes, instead of slowing down, he kicked into a higher gear. He saw an unmet need in the continuous glucose monitor (CGM) market and built the company Not Just a Patch, which sells custom adhesive patches that protect CGM devices. Through that business he’s served 400,000 patients in 65 countries, giving him incredible exposure to the T1D community.
Business was booming, but something was missing in how Lomas managed his own condition. He realized that despite his insider knowledge of the diabetes industry, he’d become gradually less engaged with his healthcare providers and his care overall. It dawned on him: if he felt disengaged, how much harder must it be for someone without his experience?
Patient disengagement is a widespread challenge in healthcare and particularly prevalent in chronic diseases like diabetes. Patients move from an initial hyper-awareness of their condition and what it’s going to take to live and thrive, and then gradually drift away from physician communication and care plans. This devolving of the healthcare relationship leads to lower health plan adherence and worse outcomes.
We’ve known this to be a problem for decades, yet most healthcare systems remain notoriously complex, bureaucratic, and functionally divided. Patients often report feeling overwhelmed by the amount of information provided — or lack thereof — and being unsure of how to approach wellness proactively. Treatment methodologies and tools are often one-size-fits all, rather than focusing on the patient’s unique challenges.
Creating a healthcare experience that engages patients, rather than leaving them confused on the sidelines, is critical to improving health because it’s the only thing that will lead to behavior change. And behavior change is often the key to taking control of a chronic condition.
Lomas considered these things and emerged with a question: How can the mundane T1D patient experience — consisting of the same questions and the same predictable approaches — be transformed or complimented to become more engaging, authentic, and personalized?
In the summer of 2022, Lomas began to have clarifying conversations with T1D experts and technologists on this topic. Market research confirmed the challenge he’d identified. What was less clear was the optimal technology to use to solve the problem. That is, until he reached out to Professor Byeong Kang.
Prof. Kang is the co-inventor of a unique alternative machine learning methodology focused on knowledge acquisition and management. This approach differentiates itself from traditional machine learning by actively emphasizing the acquisition and utilization of domain-specific knowledge. He recruited Prof. Kang to his team and the framework for Balance Health began to take shape.
Under the Hood
Let’s step into the world of Balance Health and look at an example patient encounter with Bally, the conversational assistant powered by this machine learning approach.
Meet Sarah, a Type 1 diabetes patient who has been struggling with managing her condition. Frustrated with the traditional healthcare system, she turns to Balance Health in search of a more personalized and empowering experience.
Sarah opens the Balance Health app on her smartphone and initiates a conversation with Bally, the AI-powered assistant. “Hey Bally, I’m feeling overwhelmed with my diabetes management. Can you help me?”
Bally, equipped with an extensive knowledge base curated by real human experts, engages in a natural and empathetic conversation with Sarah. It starts by asking a series of insightful questions to understand her specific challenges, recent experiences, and goals.
Based on the data from patient-provider interactions and the expertise fed manually into the AI, Bally recognizes the context of Sarah’s situation and tailors its responses accordingly. It offers guidance, educates her on relevant topics, and provides actionable insights to address her concerns.
For example, Bally might say, “Sarah, I understand managing diabetes can be overwhelming. Have you tried incorporating regular exercise into your routine? Studies have shown that physical activity can help improve blood glucose control. Let’s explore some exercise options that suit your lifestyle and preferences.”
Sarah feels a sense of relief as Bally’s conversational style and personalized recommendations resonate with her. She continues the dialogue, seeking advice on nutrition, medication management, and mental wellbeing.
As the conversation progresses, Bally draws upon its growing knowledge base, enriched by the expertise of endocrinologists, psychologists, diabetes educators, and other healthcare professionals. It seamlessly integrates their collective wisdom, delivering comprehensive guidance to Sarah in a Siri or Alexa style assistant.
Throughout the encounter, Bally provides ongoing support, reminding Sarah to check her blood glucose levels, take medications as prescribed, and offers encouragement to stay motivated on her journey. It also offers resources such as educational articles, videos, and access to support groups for further assistance.
Over time, as Sarah continues to engage with Bally, the AI learns from her experiences, gaining a deeper understanding of her unique needs and preferences. This iterative process allows Balance Health to continuously refine its recommendations and provide increasingly personalized and effective guidance.
Balance Health aspires to be a conversational interface that leverages natural language processing to simulate meaningful conversations between patients and clinicians. While most machine learning models rely on fixed datasets, Balance Health takes a different approach. It harnesses the knowledge of real human experts in various fields by manually feeding their insights into the AI system. This is achieved by carefully listening to patient and provider interactions and incorporating the wisdom gained from those exchanges into the AI’s knowledge base for enhanced, optimized subsequent encounters with the patient.
The goal is to make health tech feel less like a machine and more like a real person, with the added benefit of being able to learn from past conversations. The interface records and saves conversations to this knowledge base, allowing the technology to adapt and improve over time evolving into a buddy system type app that uses that knowledge to uniquely support them.
Balance Health is currently working through the iteration of their MVP, adapting to the swift changes in ai adding capabilities such as the Bally app and evolving their knowledge capture capabilities to be sourced through conversations live as they happen rather than the current manual input.
Last Word
Pete Lomas’s built-in network of pre-existing customers with T1D from his CGM business offers a sounding board for Balance Health to ask the right questions to their target market and continue to evolve their knowledge base. With the immense potential to reach a global market, Lomas and his team are adapting to the rapid changes in AI by taking their time to confirm their hypothesis that a knowledge acquisition/management approach to machine learning is the right answer to bridging the disengagement gap between T1D patients and healthcare providers. The development of the Bally “Buddy System” app, as part of their offering, manifests this vision. However, Balance Health isn’t just a result of technological breakthroughs; it is the embodiment of Pete Lomas’s own journey as a T1D patient, his experience in the healthcare tech sector, and his enduring pursuit of improving patient engagement.
Balance Health embodies a future where patients can reclaim control of their conditions, making healthcare feel more personal and less like ticking off boxes in a standardized process. Lomas’ vision challenges the status quo by turning the tables on conventional patient engagement strategies, pushing the boundaries of how AI can be harnessed to foster a more patient-centric healthcare model. In a world where healthcare often feels depersonalized, Balance Health stands as a beacon, illuminating a path towards an era where every patient feels truly seen, heard, and empowered in their journey towards better health.
By integrating technology with a profound understanding of patient needs, Balance Health aims to rewrite the narrative of patient disengagement, establishing a new standard for patient-centered care.
→ Connect with Balance Health via email
Call for T1D Innovation
Are you a scientist or innovator focused on T1D innovation who would benefit from education about how to navigate and build a company that will be successful in attracting mission-aligned capital, customers, and collaborators to pursue scientific discoveries in the field of Type 1 diabetes? Learn more and apply for a T1D Fellowship.
Become a Health Moonshot Champion
Health moonshots are fueled by passionate families, foundations, and industry organizations committed to achieving health moonshots. Learn how you can join other champions of the T1D Moonshot or one of our other global health moonshots.
Funders: Learn how you can become a Health Moonshot Champion and invest in Health Transformers.
Founders: Don’t make the journey alone. Learn how Health Transformer University fuels your health moonshot.
Follow us on social media for daily updates on Health Transformers: Twitter, LinkedIn, Facebook, and Instagram.
Published: Aug 3, 2023