🚨 Pulse of Innovation 🚨

Breaking news in the healthcare AI

How Blue Cross Blue Shield embraced AI and Transformed for Better

Blue Cross, Blue Shield of Michigan (BCBSM) is non-for-profit mutual health insurance company and has worked with very small profit margins with traditional slow technology infrastructure. With expansion of insurance coverage and technology rule it required a transformation to keep up with industry trends.

They successfully navigated the complexities of the healthcare industry and embraced AI technologies to improve efficiency and customer service while adhering to strict regulations. This transformation journey, led by their EVP/CIO Bill Fandrich, has been guided by seven key principles:

1. Board engagement: Fandrich regularly briefs the board on AI initiatives, ensuring they understand its potential and impact.

2. Cross-functional collaboration: A dedicated team representing various departments oversees AI projects, ensuring compliance and facilitating rapid development.

3. Secure access: BCBSM developed SecureGPT, a system that provides role-based access to LLMs and safeguards sensitive data.

4. Architecture matters: A cloud data solution integrates data from different units, streamlining operations and supporting AI applications.

5. Employee education: Ongoing training on responsible AI practices ensures employees understand the importance of ethical and compliant AI use.

6. Bias mitigation: BCBSM employs testing algorithms and training data procedures to minimize bias in AI applications.

7. Safe haven for innovation: BCBSM's subsidiary NASCO serves as a technology innovation hub, allowing for experimentation and development of new solutions outside of daily operations.

By adhering to these principles, BCBSM has become a leader in AI innovation within the healthcare sector, demonstrating that responsible AI implementation is achievable even in a highly regulated environment. Their success story serves as an inspiration for other organizations looking to leverage AI for positive impact.

🧑🏼‍🔬 Bench to Bedside👨🏽‍🔬

Developments in healthcare AI research and innovations

Digital Twin in Managing Hypertension Among People With Type 2 Diabetes

Objective:

• To explore the potential of digital twin technology in managing . hypertension among patients with Type 2 diabetes.

• Focus on creating personalized treatment plans through the use of digital twins to improve patient outcomes.

Methodology:

• The study involves developing a digital twin model that simulates the physiological state of patients with Type 2 diabetes and hypertension.

• The model integrates patient data, including clinical, lifestyle, and genetic information, to predict disease progression and treatment responses.

• The digital twin is updated in real-time with new data from the patient, allowing for continuous monitoring and adjustment of treatment strategies.

  

Results:

1. Blood Pressure Reduction: The DT group showed a more significant reduction in systolic blood pressure (SBP) by 7.6 mm Hg and diastolic blood pressure (DBP) by 4.3 mm Hg compared to the SC group, which had a reduction of 3.2 mm Hg in SBP and 2.2 mm Hg in DBP.

2. Hypertension Remission: In the DT group, 50% of participants with hypertension achieved remission after one year, while none of the participants in the SC group achieved remission.

3. Microalbuminuria: a marker for kidney damage, was reduced more significantly in the DT group.

   

Pros and cons:

Pros: 1. Innovative Approach: The use of AI-driven Digital Twin technology offers a novel and personalized approach to managing hypertension and type 2 diabetes.

2. Significant Outcomes: The study demonstrated substantial improvements in blood pressure control, reduction in anti-hypertensive medication use, and hypertension remission rates in the DT group.

3. Comprehensive Intervention: The DT intervention not only targeted BP management but also improved glycemic control, weight, and other cardiometabolic risk factors.

Cons: 1. Ethnocultural Bias: The study’s findings may not apply universally due to the specific population studied, which could limit its broader applicability.

2. Small HTN Subgroup: The relatively small number of participants in the hypertension subgroup may weaken the statistical power of the results.

3. Limited Long-term Data: The study’s one-year duration limits the ability to assess the long-term sustainability of the observed benefits Nature Scientific Article

🧑🏽‍⚕️ AI in Clinic 🏥

Developments in healthcare AI research and innovations

AI in Medical Education

Dendritic Health AI is transforming medical education with advanced AI tools. Their flagship platform, Neural Consult, uses AI-powered large language models to provide medical students and professionals with tailored healthcare education resources. The platform aims to improve diagnostic accuracy and efficiency in clinical training.

Features 

• Question generator: generate unlimited questions on any medical topic with clear explanation.

• Clinical case simulator: choose a case and simulate the experience of going through a real clinical assessment

• Lecture summaries: summarise any medical lecture into concise, easy to digest outline.

• Anki generator: generate anki decks and cards from your n notes or lecture summaries.

• Differential diagnosis: Provides accurate differential diagnosis based on interactions.

• Used by leading medical schools such as U Michigan, Northwestern, UNC and more

AI notebook app

AI Notebook seamlessly organizes text, image, audio & YouTube video on your phone, for a powerful note-taking experience that won't miss a beat. Ask questions and save them all in one place for easy access later.

1. Generate transcripts and AI summarise for College Students in lectures.

2. Specialising in YouTube Video Summariser, PDF Summariser, Article Summariser.

3. Save key insights and review with study guides, quizzes, and flashcards.

4. Build your AI-powered notes library

🤖 Patient First, AI Second🤖

Ethical and Regulatory Landscape of Healthcare AI

Training Healthcare Workforce for Generative AI for Efficient Care

• OSF HealthCare, a 16 hospital health network, facing workforce shortages and the need for efficiency, launched a mandatory generative AI education program for its 24,000 employees. The goal was to quickly educate everyone on generative AI, its potential benefits and risks, and how to use it responsibly within the organization. The program was designed to be engaging and concise, using mixed media and AI-generated content to cater to short attention spans. It also emphasized the importance of validating AI-generated results and protecting patient data.

• The education was tailored to two groups: leaders and providers, and all other roles. It was delivered through the internal Corporate University platform and included quizzes to ensure comprehension. The results were impressive, with 78.75% completion rate and positive feedback on content relevance and knowledge enhancement.

• The success of the program highlights the importance of timely AI education in a rapidly evolving technological landscape. Key takeaways include understanding the audience, using innovative approaches, and leveraging AI tools for content creation. OSF also plans to continue evolving its AI education efforts through initiatives like Prompt Buddy, a platform for sharing helpful AI prompts.

• OSF emphasizes the importance of critical thinking and fact-checking to mitigate the risks associated with Al's potential for inaccuracies. Additionally, the training reinforces the organization's commitment to patient privacy and data security by outlining strict guidelines for handling sensitive information.

• Training healthcare workforce on generative AI is critical to ensure patient safety, team member efficiency and overall program success.

READ FULL ARTICLE HERE