By Daniel Schwen

The 3 most-cited studies in healthcare and AI

Carlos Folgar and Jess McCuan


Is there a robot doctor in the house?

Artificial intelligence is a hot topic in many industries, but particularly healthcare, where doctors and patients are now looking to AI-based apps for help with everything from diagnosing minor ailments to sizing up heart problems. In fact, AI seems set to transform several corners of health and medicine, from cancer prognoses to genomics.

At Quid, we analyzed the latest academic research to see exactly where AI is making an impact in healthcare. For this we used data from SCOPUS -- the largest abstract and citation database of peer-reviewed academic literature from around the globe. In just a few minutes, Quid sifted through 2,000 recent academic papers about AI in healthcare.

The software created the map above, extracting focus area trends from recent research. The top left region shows studies of human psychology and cognition, such as active learning, social media on psychology, and game studies. The top middle focuses on monitoring people, specifically their motions, emotions, and health signals for the elderly. The right side involves applying advanced algorithms to diagnose different conditions. Squarely in the middle of the map: studies focused on IoT, digital health, and predictive modeling; their central location and dispersed connectivity to other parts of the map tell us they are highly influential in the entire landscape of AI in healthcare.

Next, we wanted to see which areas of healthcare were mentioned most frequently in papers about AI.

Predictive modeling of clinical data (light purple) and clinical decision support systems (light red) show up frequently in papers about cancer and diabetes, compared to dementia & alzheimer’s. Text & semantic analysis (dark purple) is being used more often in cancer research, compared to the other two therapeutic areas. Support vector machine classifying models (dark red) and artificial neural networks (peach) are being equally applied across all three therapeutic areas.

Then, we saw that citation metadata from SCOPUS indicates the papers below have already been cited elsewhere, a sign that they are being quickly validated by others in the scientific community.

The top 3 cited papers:

  • Social big data: recent achievements and new challenges

    • Presents a revision of new methodologies generated by big data technologies and machine learning algorithms to allow for efficient data mining and information fusion from social media and psychological study data.

    • Affiliations: Computer Science Department, Universidad Autónoma de Madrid, Spain; Department of Computer Engineering, Chung-Ang University, Seoul, South Korea

  • Deep Convolutional Neural Networks for Computer-Aided Detection

    • Reveals three previously understudied factors of employing deep convolutional neural networks (CNNs) to computer-aided detection problems: different CNN architectures, the influence of dataset scale and spatial image context on performance, and the when and why transfer learning from pre-trained ImageNet can be useful. The findings can be extended to the design of high-performance CAD systems for multiple medical imaging tasks.

    • Affiliations: National Institutes of Health Clinical Center, Bethesda, MD, United States

  • Transition-aware human activity recognition (TAHAR) using smartphones

    • For a system that targets real-time classification with a collection of inertial sensors, this research proposes two implementations that differ in their prediction technique as they deal with transitions either by directly learning them or by considering them as unknown activities. Results show that TAHAR outperforms state-of-the-art baseline works and reveal the main advantages of the architecture.

    • Affiliations: University of Genova, Via Opera Pia 13, Genova, Italy; Universitat Politécnica de Catalunya, Rambla de l'Exposició, 59-69, Vilanova i la Geltrú, Spain

Because healthcare is so highly regulated, it may take months or years for treatment-related research to make its way into your doctor’s office. But the Quid map shows us plenty of technology in the diagnostic software and digital health fields that could be easily commercialized. 

And while plenty of people remain skeptical that machines could ever fully replace doctors, a report out this week suggests AI is rapidly gaining ground

Keywords: Ai, Healthcare, Medicine, Device

Intelligence in your Inbox

Sign up for the Quid newsletter for a monthly look into how data and visualization are changing the way we view the world.

Get to know Quid. Chat with us for about 15 minutes about your project, and we'll create a free guided test drive to help you find solutions.

clear x