Telemedicine healthcare doctor

Accelerating Healthcare Diagnostics with Intel oneAPI and AI Tools



A 2020 survey 1of US healthcare leaders by Intel showed that 84% of respondents started or planned to start using AI following the onset of COVID-19, a nearly twofold increase from the months before COVID-19. Ninety-four percent of respondents using AI targeted predictive analytics for early intervention.

The promise of AI in healthcare and life sciences is profound. It has the potential to help clinicians and researchers prevent disease, speed recovery, and save lives by unlocking complex data. It can also free them from mundane tasks so they can focus on their patients or research.

AI in medicine, pharmaceutical research, and other areas of healthcare can help improve patient care as well as overall population health.2 Today, deep learning and machine learning in healthcare are streamlining workloads for clinicians, informing personalized treatment plans, and enhancing patient experiences.

Using AI to speed diagnostics for individuals or larger populations involves complex computing models that benefit from accelerated computing systems built on a mix of CPUs, GPUs, and other specialty processors. Over the past two years since the onset of COVID-19, several healthcare organizations and technology providers have found that Intel oneAPI and AI tools efficiently accelerate applications requiring high-performance, multiarchitecture computing.

GE Healthcare* Solutions Accelerated by Intel® oneAPI Toolkits

GE Healthcare* collaborated with Intelon oneAPI to make its code cross-architecture ready. Intel® oneAPI Toolkits along with Intel® Xeon® processors are used extensively to drive their medical devices like X-ray machines, mammography scanners, CT, MR, pad, and molecular imaging scanners. When combined with Intel oneAPI tools, the Intel oneAPI Math Kernel Library helped boost performance. For AI optimizations, porting CUDA* Deep Neural Network (CuDNN) library to Intel® oneAPI Deep Neural Network Library was straightforward. As a result, GE Healthcare estimates saving potentially millions of dollars in configuration costs and years of engineering efforts. Going forward GE Healthcare is considering training their scientists to start using oneAPI and DPC++ for their research and development activities, which will help them to run their prototype code, take advantage of CPUs, and accelerate using GPU device making code more efficient.

Mental Health Diagnosis: Hipposcreen

Globally, an estimated 5% of adults suffer from depression4, a prevalent and serious illness that, if left untreated, can become increasingly debilitating even to the point of suicide. Unlike most other psychiatric diagnoses, there is no one-size-fits-all diagnostic procedure or equipment for depression itself, and while there are some cases that can be clinically diagnosed, most assessments are dependent on subjective self-descriptions by the patients themselves. More problematic still, due to the stigma surrounding depression, it may take months or even years for sufferers to figure out that their somatic symptoms are caused by this major depressive disorder. 

HippoScreen Neurotech Corp. is a Taiwan-based startup company backed by Compal Electronics, the second largest contract laptop manufacturer in the world. With electroencephalogram (EEG) signal processing and artificial intelligence technology as the pillars, HippoScreen is developing EEG-assisted medical diagnosis tools. HippoScreen has collaborated with three medical centers in Taiwan to build the largest cross-center EEG-based database on clinical depression in the world.

HippoScreen used tools and frameworks in the Intel® oneAPI Base and AI Analytics Toolkit5 to the improve efficiency and build times of deep learning models used in its Brain Waves AI system. As a result, HippoScreen can broaden the system’s applications to a wider range of psychiatric conditions and diseases.

According to Daniel Weng, HippoScreen CTO, “We at HippoScreen have been able to take advantage of the software optimizations in Intel® Extension for Scikit-learn* and Intel® Optimization for PyTorch* to accelerate the build times for the AI models in our customized EEG Brain Waves analysis system by 2.4x.” Intel® VTune™ Profiler helps accelerate Hippo Screen’s AI-based NeuroTech solution for mental health diagnosis and treatment by 2x.

Cancer Detection: KFBIO

Cervical cancer is the fourth most frequent cancer in women with an estimated 570,000 new cases in 2018, which represents 6.6% of all female cancers.6 Early diagnosis through Papanicolaou (Pap) screening along with treatment significantly improve the chances of patient survival. The liquid-based cytology (LBC) method for cervical cancer screening is used by over 90 percent of the Pap tests performed in the United States, according to the National Institutes of Health (NIH)7.

Ningbo Konfoong Bioinformation Tech Co., Ltd (KFBIO) is a total solution provider of pathology related products, which include sample processing equipment, digital pathology scanning systems, pathology information system and deep learning algorithms and models for detecting and classifying precancerous changes and abnormalities. Building upon its core strength as a leading medical pathology slide scanner manufacturer in China (thousands of devices deployed in various hospitals), KFBIO’s capability in scanning traditional pathological sections into digital images allows scientists to apply deep learning techniques to assist medical diagnosis.

Using Intel® Optimization for TensorFlow* and Intel® Distribution of OpenVINO™ toolkit, KFBIO increased inference performance of cervical cancer screening by up to 8.4X on Intel® Xeon® Gold 6148 processors. KFBIO is also a member of the Intel® AI Builders program8, an ecosystem of industry leading independent software vendors (ISV), system integrators (SI), original equipment manufacturers (OEM), and enterprise end users, who have a shared mission to accelerate the adoption of AI across Intel platforms.

Lung Disease Detection: Accrad

After chest X-rays are taken to detect COVID-19, viral pneumonia, and other diseases, days or weeks may pass before a radiologist reviews the images. Such delays can worsen patient outcomes and extend outbreaks of contagious diseases. 

Accrad, a medical AI software company based in South Africa, has developed an AI-powered solution called CheXRad, which is capable of labeling certain pathologies in chest radiographs up to 160x faster than radiologists, and at comparable levels of accuracy, sensitivity, and specificity.9

With the help of the Intel® oneAPI AI Analytics Toolkit, Intel Distribution of OpenVINO toolkit, and Intel® Developer Cloud for oneAPI, Accrad was “able to train, optimize, and deploy a machine learning model in less time and at a lower operational cost than available alternatives, enabling us to get to market fast with a powerful solution that’s optimized for Intel architecture.”