Transforming healthcare through the power of Generative AI

Generative AI is poised to reshape the future of healthcare, driving innovations in patient care, diagnostics, and operational workflows. In an exclusive interview with Express Computer, Prakash KS, Head of Gen-AI CoC at Siemens Healthineers Global Development Center, shares the strategic vision behind these transformative technologies. From integrating multimodal data for better decision support to enhancing remote health capabilities, Prakash discusses how Siemens Healthineers is leveraging AI to create a seamless, patient-centred ecosystem, while addressing pressing challenges like data security and global healthcare accessibility.

What is the strategic vision for generative AI at Siemens Healthineers, and how do you see it transforming the healthcare space?

Siemens Healthineers fundamentally looks at healthcare as an end-to-end delivery model. It’s not just about the physician; it’s about patient well-being, the patient’s family, and the physician as a touchpoint. Our approach is focused on impacting all three areas.

We first aim to make a direct impact on patient care. For instance, a use-case is the utilisation of AI systems to take up mundane tasks from doctors so that they are able to give their patients more time.

Next, we consolidate data from multiple sources – imaging devices, EMR systems, and diagnostic reports – that present data in different formats. Using technology, we can bring this data together and enable decision support for physicians.

From the patient’s perspective, we aim to optimise workflows. Today, navigating a care facility from diagnosis to treatment can be tedious. Can we simplify this process, eliminate repetitive steps, or even predict symptomatic matches for conditions before they manifest?

Another focus is remote health. Can we enable remote diagnosis and predictive care at a single touchpoint? AI can also enhance reporting, providing simplified summaries that help families understand a disease, its care process, and its implications.

Care doesn’t always equate to a solution sometimes, care is ongoing. The strategy of Siemens Healthineers is built around better reporting, integrating data from imaging and other sources like genomics, and creating seamless workflows. Auxiliary goals include ensuring equipment reliability and improving user training for healthcare professionals.

You mentioned patient care and safety as key priorities. Are there specific GenAI or AI case studies at Siemens Healthineers that highlight these efforts?

We’ve worked on several initiatives. One example we demonstrated at last year’s Radiological Society of North America (RSNA) congress, involved integrating diagnostic imaging with reporting. This combines structured data with unstructured data, providing actionable insights for care providers. We are also working on generative AI for use in customer service and support as well as for training medical staff. 

Another initiative is clinical decision support, where we analyse data from blood tests, clinical data (age, gender, etc.), and other sources to identify potential susceptibility to diseases. While not definitive, AI can assist clinicians by pointing out areas to monitor. This is still human-in-the-loop, meaning the AI supports but doesn’t make decisions.

Data security is critical, especially given rising security threats. How does Siemens Healthineers ensure data safety?

Data security involves three aspects: collection, storage, and persistence, all while adhering to legal and highest regulatory standards.

First, we ensure clarity and transparency in data collection, following GDPR timelines and offering users control over their data. This includes allowing data deletion when required.

Second, we practice data economy by using only the data necessary for algorithm development, avoiding overengineering.

Third, we secure data through robust authentication, authorisation, and need-to-know access. We comply with regulatory frameworks globally and aim to stay ahead of emerging standards. Additionally, patient data is anonymised at the source unless explicitly required for a specific use case.

Looking ahead, what trends or innovations do you see shaping healthcare in the coming years?

Healthcare will become more precise and personal. Two key enablers are cloud computing and advancements in communication technologies.

Over a decade ago, we recognised that devices needed connectivity, even before fully understanding its implications. Today, connectivity enables AI-driven precision medicine, including drug discovery, robotic surgeries, and remote healthcare solutions.

This is also driven by a global shortage of healthcare workers, especially in developed and underserved regions. Remote health technologies will address these gaps, enabling expertise to be delivered where needed.

In the future of medicine, digital patient twins will help in early diagnostics and precision care and Siemens Healthineers is working on developing the same. Powerful AI-enabled applications are already in use that contribute to the concept of the digital patient twin. Using models based on algorithms and large amounts of data, we try to better understand people’s health status in order to improve treatment outcomes. It is a long-term vision, but it is already technically feasible today in partial solutions. Although technology exists, the implementation of these concepts in clinical application will take many years. Hence, it is important to create the conditions in hospitals today and to continuously implement the available partial solutions.

Additionally, the integration of AI in multi-modality imaging decision support has significantly enhanced healthcare in India. It has expedited diagnoses by providing accurate insights from various imaging sources like MRI, CT scans, and X-rays, aiding in early disease detection and treatment planning. This technology also optimises resource utilisation and reduces costs. In India, where healthcare resources can be scarce, AI-based decision support systems extend access to quality care in remote areas, enable telemedicine, and support overburdened healthcare professionals.

Security is also a concern for remote health. Do we need new measures to address these risks?

Security must be proactive, not reactive. Historically, security measures were developed in response to breaches. That approach no longer suffices.

Today, we focus on end-to-end security, from patient devices to communication channels. Emerging solutions like federated learning where data is processed locally rather than transferred can enhance security. With the advent of 5G and future technologies like 6G, we expect these measures to evolve further.

What’s your IT roadmap for the healthcare industry over the next few years?

Healthcare IT will focus on precision medicine and personal care. Patient touchpoints will become more accurate, improving image diagnosis, automation, and post-processing. IT systems will enhance communication, providing concise reporting and mapping historical data to current diseases.

Remote health and robotic surgeries will gain traction. Additionally, drug discovery and management will evolve to enable real-time monitoring and adjustments. Home health is another area to watch.