Microsoft’s AI Magic: Transforming Healthcare Insights

Microsoft introduces cutting-edge data and AI solutions that revolutionize healthcare, offering unprecedented insights and enhancing the patient experience

Every sector needs unique insights to succeed, and harnessing data’s value is vital. In healthcare, data may improve health outcomes, patient and clinician experiences, and organisational performance. The appropriate data strategy may change lives when patient or population health is at the centre of decision-making.

Organisations are battling to keep up with rising healthcare data volume, diversity, and velocity. The World Economic Forum estimates that hospitals generate 50 petabytes of segregated data every year, or 10 billion music files. As much as 97% of this data is wasted, preventing useful discoveries. Utilising all this data may lead to clinical and operational innovations that improve patients’ healthcare experiences. As organisations realise that without a sound data strategy, they are simply scratching the surface of AI’s potential, data becomes more important.

Microsoft are showcasing new data and AI technologies and capabilities at HLTH 2023 to assist healthcare organisations improve patient and clinician experiences while providing excellent care more effectively and cheaply. These new solutions allow healthcare organisations to use Microsoft Cloud for Healthcare’s breadth and scalability with a cohesive, secure, and responsible data and AI strategy.

Combining data analytics to improve patient care and corporate value

Microsoft introduced Microsoft Fabric in May, an end-to-end, unified analytics platform that combines all the data and analytics capabilities organisations need to harness their data’s potential and prepare for AI. Today, Microsoft Cloud for Healthcare powers the first industry-specific data solutions in Fabric that combine data and insights via one architecture and experience. The healthcare data solutions in Fabric, now in preview, eliminate the costly, time-consuming process of stitching together a complex set of disconnected, multimodal health data sources (text, images, video, etc.) and provide a secure, governed way for organisations to access, analyse, and visualise data-driven insights across their organisation

MS Fabric offers healthcare organisations:

• The capacity to integrate data from EHRs, PACS, laboratories, claims, and medical devices throughout an organisation. With open data standards integrating FHIR, DICOM, and MedTech services, the Fabric data lake integrates structured, unstructured, image, and medical device data into one architecture.
• Connectors and converters simplify FHIR, DICOM, and MedTech data conversion and pipeline creation for particular use cases.
• A multimodal data foundation that enables standardised, scalable solutions to speed clinical and operational insights and improve patient care. Fabric lets health data exist in a single data estate for AI model building and analytics.
• Clinical research is enabled by OMOP analytics, and patient outreach analytics personalise patient engagement.
• Using machine learning models to extract, redact, or surrogate identifiers from unstructured data like doctor’s notes, medical documents, and clinical trial studies, a new de-identification service will help organisations protect patient-protected health information (PHI).
• Microsoft preview healthcare-specific categorization rules, labels, and data glossaries help healthcare organisations control, secure, and manage their data estate.

Early adopters of Microsoft Fabric expect to utilise the analytics platform to advance some of their most important healthcare use cases:

• Northwestern Medicine, Chicago’s premier integrated academic health system, will use Fabric’s healthcare data solutions to integrate clinical data from multiple sources, meet regulatory information exchange requirements, and unlock insights with data and AI to advance their patients-first mission with high-quality and timely care.
• Arthur Health and Quisitive for the Ontario Workers Network will utilise Fabric to build predictive care stage models. A statewide network of hospitals, including Ottawa Hospital, OWN has skilled physicians that offer world-class treatment for employees in their communities.
• SingHealth, Singapore’s biggest public healthcare network, wants Fabric’s healthcare data solutions for its data infrastructure. This will improve healthcare delivery and empower individuals to manage their own health.
• Microsoft Fabric helps solve healthcare’s disparate data challenge. Fabric’s healthcare data solutions are explained here.

New AI tools empower patients and demystify medical terminology

New healthcare capabilities in Azure AI services will help organisations maximise AI’s value to improve patient outcomes:

• Giving doctors and researchers informed choices Azure AI Health Insights is a cognitive service that delivers prebuilt models that analyse and draw conclusions for physicians and academics to employ in critical patient care situations.

Microsoft are previewing three new models, including patient timeline, which uses generative AI to extract key events from unstructured data like medications, diagnosis, and procedures and organise them chronologically to give clinicians a more accurate medical history to better inform care plans. Clinical report simplification employs generative AI to simplify medical jargon while maintaining the complete clinical information for sharing with patients. Radiology insights verifies quality by reporting mistakes and discrepancies. The programme finds follow-up suggestions and clinical findings in clinical record with radiologist-measured sizes.

• Healthcare chatbots and virtual assistants with generative AI This new preview capability in Azure AI Health Bot provides out-of-the-box healthcare intelligence that can be customized and integrated into existing workflows, using answers from a healthcare organization’s content sources and generative AI to provide answers from credible sources like the National Institutes of Health and the FDA.
• Medical data extraction and labelling for insights Text Analytics for health, an Azure AI Language service, extracts and labels medical data from unstructured texts using machine learning. Population health, Azure OpenAI Service patient population Q&A, clinical trial patient cohorts, and bulk historic data processing are new industry open source templates.
  

AI helps physicians provide personalised, high-quality treatment

To reduce administrative load and clinician fatigue, which grew to 53% among doctors in 2023 from 42% in 2018, U.S. health institutions are using AI-powered solutions, according to Medscape. To reduce clinician fatigue and enable personalised treatment, Dragon Ambient eXperience (DAX) Copilot, previously DAX Express, is now available for general use.

DAX Copilot, part of the Nuance Dragon family of products utilised by over 550,000 users globally, lets physicians automatically and securely draught clinical summaries in seconds from exam room or telehealth interactions for EHR review and submission.

Atrium Health was the first to implement Nuance DAX Copilot and expects to licence further throughout its 40 hospitals and 1,400 care sites. Physicians claim significant time savings in patient visit documentation. Medical staff at Atrium Health say this enhanced documentation system saves them up to 40 minutes every day. Also, 68% say their caregiving experience has improved.

Microsoft Cloud for Healthcare in AI/data age

Microsoft Cloud for Healthcare adds industry-relevant data solutions, application templates, and AI services to the Microsoft Cloud, helping healthcare organisations speed their data and AI journey. An unequalled worldwide network of trustworthy partners may customise our products. To provide holistic solutions for healthcare clients’ particular business concerns, Microsoft partner with premier ISVs and system integrators.

Microsoft healthcare solutions are based on trust and Microsoft’s Responsible AI. These innovations help healthcare organisations create connected experiences at every point of care, provide collaboration tools, empower the healthcare workforce, and unlock the value of clinical and operational data using industry-relevant data standards.

News source: