Uncovering AI’s Influence on Healthcare with Michael O’Connell

David Skerrett, Drum Digerati and digital transformation expert, kicked off our webinar and got our audience up to speed on the state of AI in the healthcare sector today.

Firstly, David took a look at the history of hospitals. His research suggests that hospitals have been around since the 9th century, and have gone through four revolutions. But it wasn’t until the introduction of sanitation in the 19th century that the quality of healthcare dramatically improved and life expectancy started increasing. The 20th century saw a further boom in life expectancy with the invention of antibiotics and vaccines.

Now, we’re seeing a new revolution in healthcare, one powered by digital technology and, in particular, analytics and AI. On the one hand, life expectancy will continue to see improvements. But some of the most impressive statistics come in where cost efficiencies are concerned. For instance, David shared that AI applications can create $150 billion in annual savings for the United States’ healthcare economy by 2026. Meanwhile, the top three applications that represent the greatest near-term value are robot-assisted surgery ($40 billion), virtual nursing assistants ($20 billion) and administrative workflow assistance ($18 billion). With savings and new value creation like this, healthcare organisations will be able to redeploy efforts and treat patients more efficiently.

There are other areas that David highlighted as a possible beneficiary of AI/ML: chronic health conditions. Research suggests that AI/ML will help to develop new treatments for cancer (63%), heart disease (63%) and diabetes (66%). But we don’t have to wait for new AI solutions to see the benefits the technology is already having on the healthcare sector. David discussed how Diagnosis, Treatment, Drug Design, Predictions & Efficiency, and Patient Care are all benefitting from AI solutions today.

• Diagnosis: Misdiagnosing illness and medical error accounted for 10% of all US deaths in 2015. AI tools can rapidly identify patterns in big data to diagnose accurately.
• Treatment: AI can personalise treatments to the patient, providing unparalleled insights into how a drug is performing in real-time. If a better treatment plan is available, the solution will recommend it.
• Drug Design: With getting new drugs to human trial extremely expensive and time consuming, AI solutions are helping to identify the most suitable drug choices and save time and money.
• Predictions & Efficiency: Predictive analytics is helping healthcare managers to optimise business operations and patient care through optimising staffing levels, increasing utilisation, and predicting demand.
• Patient Care: AI-powered chatbots are helping to take the strain off call centres by answering patient questions autonomously. Some organisations have even deployed chatbots to help in the coronavirus pandemic.

After David had finished setting the scene, we moved to Michael’s section of the webinar, an in-depth look at a handful of major use cases for AI in healthcare. Drawing on his experience working with US healthcare providers as part of his role as Chief Analytics Officer at TIBCO Software, Michael first discussed drug design before looking at predictions and efficiency improvements.

To start our tour of TIBCO’s AI use cases in healthcare, Michael looked at translational medicine and high content screening. For a detailed look into what Michael unveiled, view the full webinar now.

In Michael’s translational medicine example, he highlighted how AI and ML solutions accurately predict clinical outcomes for potential drugs. These solutions are sequencing molecular states and identifying the small number of clinically significant molecular states from millions of features. It works by identifying common biomarkers between new molecules and existing drugs which have vast amounts of data to work with. Established patterns and relationships reveal the potential survival rates, risk scores and novel treatment plans of new drugs.

Michael then turned his attention to high content screening, where automated microscopes capture millions of images of cells in a variety of conditions to analyse drug effectiveness. As he showed, the data analysis workflow of today can sometimes be very convoluted. It must go through six stages: Automated Image Capture; Image Pre-Processing; Segmentation; Feature Extraction; Feature Reduction & Aggregation; Visualisation & Interpretation. TIBCO’s Spotfire solution streamlines this approach. Its deep learning algorithm has been trained on whole images, not individual objects, so segmentation isn’t needed, bringing AI automation to screening and addressing bottlenecks.

But perhaps Michael’s most exciting example of AI use within the healthcare sector came when looking at Operation Room (OR) efficiencies. You may not know this, but the OR is the most expensive minute in a hospital. Every minute that an OR stands idle, it costs the hospital $625 on average. That’s money that could be better spent actively saving lives by improving the efficiency of the hospital and its OR.

TIBCO’s Spotfire analytics solution enables the optimisation of surgery operations including on-time starts, scheduling accuracy, and block utilisation. When used by Saint Francis Hospital and Medical Center, measurable results included a 25% improvement in first case on-time starts, a 20% increase in surgery block utilisation and an estimated $450,000 value of surgery time created. Ultimately, the solution has led to better time allocation and more patients being treated, all while generating cost savings.

Elsewhere, TIBCO’s software has also helped to identify which patients will suffer post-surgery infections. By predicting which patients are most at risk, the software has allowed healthcare professionals to take a proactive approach to infection, shielding vulnerable patients and reducing the likelihood of hospital readmissions due to on-site infection – which cost US hospitals $45 billion annually.

To truly understand how TIBCO has brought about real positive change to the healthcare sector with its AI solutions, download the full webinar for free now.

AI & Empathy Take on Covid-19

In early April 2020, Stanford professor and HAI co-director Dr Fei-Fei Li presented a concept for an AI-powered in-home system that could track a resident’s health, including signs of Covid-19 – all while meeting privacy standards. The system would use cameras and depth, thermal, and wearable sensors to monitor the user’s health.