Computer Vision in Healthcare

Last time, we talked about computer vision in the fintech industry. Today, we want to concentrate on various applications of this amazing technology in healthcare. In fact, computer vision is without a shadow of a doubt the next milestone in the history of healthcare. More accurate and quicker diagnoses heightened medical process, improved medical imaging, enhanced surgery–all of that is waiting for your company thanks to artificial intelligence and computer vision. Let’s see how applications of computer vision solutions in healthcare can transform your company.

Actually, images play a crucial role in healthcare. It is estimated that images count for up to 90% of all medical data we possess[1]! Moreover, each patient’s image collection can contain up to 250GB of data! This means that physicians and analysts have a lot to do, as the image is an unstructured type of data, relatively difficult, and time-consuming to analyze.

That’s why modern healthcare institutions struggle with a lack of radiologists who could analyze all these images and make accurate diagnoses. Perhaps, that’s also a case in your clinic or hospital In the United Kingdom alone, three-quarters of clinical radiology directors say they do not have enough radiology consultants to deliver safe and effective patient care. Only one-in-five UK trusts and health boards have enough interventional radiologists to run a safe 24/7 service to perform urgent procedures[2].

If you run a hospital or a clinic, you surely understand the urgency and complexity of this problem. It’s easy to imagine that the consequences can be severe. This situation begs the question, who or what could replace these radiologists? The answer is quite straightforward–Artificial Intelligence. And that’s the very first application of computer vision in healthcare that we want to tackle.

Computer vision in healthcare helps in making diagnoses

Computer vision systems offer precise diagnoses minimizing false positives. In many instances, computer vision algorithms can be even more effective than human physicians. That’s because these algorithms are trained on thousands of medical images presenting a given disease or anomaly. As a result, AI algorithms can spot any irregularities with amazing precision. Not to mention that they are never tired, can work 24/7, never go on vacation, or sick leave.

Computer vision in healthcare can significantly improve many fields of modern medicine, i.a.:

• X-ray radiography
• Magnetic resonance imaging (MRI)
• Ultrasound
• Endoscopy
• Thermography

Thanks to the fact that computer vision algorithms are trained using a vast amount of training data, computer vision algorithms can detect even the slightest presence of an anomaly that may be missed out by human analysts. The use of computer vision in healthcare diagnosis can provide high levels of precision. And that happens even today!

Consider LYmph Node Assistant (LYNA). It’s one of the most prominent deep learning models in the computer vision field, developed at MIT. LYNA reviews sample slides and recognize characters of tumors and metastases in a short timespan with a mind-boggling 99% rate of accuracy[3]. And the good news is, your company can implement a similar model! But let’s take a look at some more examples.

ICAHN school of medicine at Mount Sinai

One of their projects was to identify markers of acute neurological illnesses, such as hemorrhages and strokes. The organization used 37,236 head CT scans from across their health systems to train a deep neural network and teach it how to determine if an image showing an acute neurological illness.

What results did they achieve? Their systems can now identify a problem from a CT scan in 1.2 seconds[4]! That’s 150 times faster than it would take a physician to read the image. In many instances, it can be a life-saving difference.

Mount Sinai has been in the news lately again, this time due to their offense against the COVID-19 pandemic. According to their website[5], Mount Sinai researchers are the first in the United States to use AI combined with medical imaging, and clinical data to analyze patients with coronavirus disease. They have developed a unique algorithm that can rapidly detect COVID-19 based on how lung disease looks in computed tomography (CT) of the chest, combined with patient information including symptoms, age, bloodwork, and possible contact with someone infected by the virus.

Again, results are more than promising. Their algorithm had a statistically higher sensitivity than human radiologists (84% compared to 75%). The AI system also improved the detection of COVID-19-positive patients who had negative CT scans.

Microsoft Inner Eye[6]

Microsoft offers Inner Eye software, which can visually identify and display possible tumors and other anomalies in the X-ray images. It’s an AI-fueled software, designed to help radiologists in their work. What does it look like? The radiologists can upload the three-dimensional patient scans into the system. The Inner Eye algorithms analyze the picture and color areas that potentially contain tumors or other anomalies.

Their software is designed to assist radiologists, not replace them altogether. If you’re thinking about ways to help your radiologists, such software can be invaluable support.

Although Microsoft doesn’t brag about the clinics and hospitals using their software, we already know that Inner Eye is FDA-approved[7], which means it can be used in hospitals and clinics throughout the United States.

Computer vision in healthcare enhances surgery

Artificial Intelligence plays a more and more important role in modern surgery, especially in Minimally Invasive Surgery (MIS). Today, operating theatres are filled with robot assistants that improve the surgeons’ work. The 3D, high-definition imaging that medical robots use, increases the vision of the operation field and makes depth perception accessible. As a result, surgical operations are more accurate and take less time.

Modern robotic surgical systems include a camera arm and mechanical arms with surgical instruments attached to them. The surgeon has control over the arms and operates them via a console placed near the operating table. The console gives the surgeon a high-definition, magnified, 3D view of the surgical site[8].

Again, let’s take a look at some real-life applications.

Gauss Surgical (Triton)

Gauss Surgical is a US-based company that offers software called Triton, which can help physicians monitor surgical blood loss using computer vision in healthcare. Triton measures the given patient’s current blood loss and the rate at which they are losing blood.

In mid-2019, this company informed the market about a new study published in the International Journal of Obstetric Anesthesia. It concluded that the use of the company’s Triton system could significantly help in monitoring maternal blood loss during labor. Researchers used the Triton system to monitor blood loss in all deliveries (3,800+) at Mount Sinai Hospital from August 2017 through January 2018.

What were the results?

• Improved Hemorrhage Recognition: Vaginal: 2.2% (Triton) vs. 0.5% (control).
• Improved Patient Management: Decrease in patients who required a transfusion outside of the labor floor, 71% (control) vs. 47% (Triton).
• Cost Savings: The annualized cost savings totaled 209,228 USD.

RSIP vision[9]

It’s a California-based company that provides computer vision in healthcare and image processing for cardiology, pulmonology, ophthalmology, orthopedics, and radiology. Among many fields, they are concentrated on orthopedics. They offer computer vision solutions that enable accurate navigation during orthopedic surgery. Their image processing techniques provide the surgeon with a highly accurate and effective real-time in-op view of the surgery environment.

Also, their systems can be used in:

• Detection of bone cancer
• Point and surface registration
• Bone segmentation

And many more.

All of the aforementioned solutions and applications are extremely helpful for clinics, hospitals, and other entities operating patients. We would like to show you one more application, which should be particularly interesting for pharmaceutical companies. As it turns out, computer vision is a chance for substantial savings, both time and money-wise.

Computer vision in healthcare improves clinical trials

Computer vision in healthcare can help in reducing the number of people who drop out of clinical trials. This occurrence is known as attrition, and it can cause some severe problems for drug development companies. As you know, every pharmaceutical company has to conduct preclinical tests before conducting a clinical test on a group of people. The company has to provide detailed information on the dosing and toxicity levels of a newly developed drug. At this point, the most important aspect is to establish if the drug is safe to use.

The next step is clinical research and trials on a group of human candidates. This part is especially important and tricky. First of all, clinical trials are the most expensive drug development stage, just to mention the need to pay every patient taking part in the tests. One of the key factors to conduct successful clinical tests is to find suitable candidates. The candidates have to represent many (preferably all) races, ethnicities, ages, and genders. And what’s more, all of the patients taking part in the tests have to be thoroughly examined before they are allowed to participate in the testing.

Many candidates are rejected before the clinical trial starts. It’s a time-consuming and costly process. That’s why pharmaceutical companies constantly think of ways of improving it. Again, AI comes to the rescue. The AI applications can help to select proper candidates and do even more. The clinical test stage is essential. Tests need to be performed in an appropriate and monitored way; only then they can be deemed successful.

AI Cure

It’s a US-based company that aims at reducing risk in clinical research. AI Cure is developing an advanced mobile platform by combining the most recent advances in artificial intelligence through deep learning, computer vision, and machine learning. Their algorithms help researchers monitor a given patient’s adherence to the prescribed treatment using computer vision.

AI Cure software is based on a mobile app that monitors patients as they undergo treatment. Currently, AI Cure is testing a new set of tools that help researchers and clinicians understand and predict how patients respond to treatment. Their algorithms capture the physiological data points that characterize expressivity, psychomotor function, and cognition.

On April 28, 2020, they published a press release where they introduced their digital biomarker platform used to detect subtle changes in a patient’s condition, fully remotely. As the company reports, “The platform leverages computer vision and AI to gather and analyze visual and auditory cues directly through the patient’s smartphone camera, pinpointing critical patient responses and behavioral trends with the frequency and accuracy needed to elevate the integrity of clinical trial data.”[10]

Today, the global healthcare sector is changing rapidly. Modern cutting-edge solutions and technologies improve the way we are treated, operated, and monitored. Computer vision in healthcare is one of many complex technologies that transform this and numerous other sectors and industries. If you’d like to find out how your organization can benefit from computer vision and other AI-related technologies and solutions – get in touch with us! We are waiting for your call!

References

[1] Heather Landi. IBM Unveils Watson-Powered Imaging Solutions at RSNA. Dec 1, 2016. URL: https://www.hcinnovationgroup.com/population-health-management/news/13027814/ibm-unveils-watsonpowered-imaging-solutions-at-rsna Accessed Jun 29, 2020.
[2] Rcr.ac.uk. The NHS does not have enough radiologists to keep patients safe, say three-in-four hospital imaging bosses. Apr 4, 2019. URL: https://www.rcr.ac.uk/posts/nhs-does-not-have-enough-radiologists-keep-patients-safe-say-three-four-hospital-imaging. Accessed Jun 29, 2020.
[3] Volodymyr Bilyk. Computer vision opportunities in Medical Imaging Explained. May 6, 2020. URL: https://volodymyrbilyk.medium.com/computer-vision-opportunities-in-medical-imaging-explained-9e046f9e2d88. Accessed Jun 29, 2020.
[4] Jen A. Miller. Computer Vision in Healthcare: What It Can Offer Providers. Jan 30, 2019. URL: https://healthtechmagazine.net/article/2019/01/computer-vision-healthcare-what-it-can-offer-providers-perfcon. Accessed Jun 29, 2020..
[5] Zahi A Fayad. Mount Sinai First in U.S. to Use Artificial Intelligence to Analyze Coronavirus (COVID-19) Patients. May 19, 2020. URL: https://www.mountsinai.org/about/newsroom/2020/mount-sinai-first-in-us-to-use-artificial-intelligence-to-analyze-coronavirus-covid19-patients-pr. Accessed Jun 29, 2020.
[6] Microsoft.com. Project InnerEye – Democratizing Medical Imaging AI. URL: https://www.microsoft.com/en-us/research/project/medical-image-analysis/. Accessed Jun 29, 2020.
[7] Marcus Roth. Computer Vision in Healthcare – Current Applications. Aug 5, 2019. URL: https://emerj.com/ai-sector-overviews/computer-vision-healthcare-current-applications/. Accessed Jun 29, 2020.
[8] Mayo Clinic. Robotic surgery. URL: https://www.mayoclinic.org/tests-procedures/robotic-surgery/about/pac-20394974. Accessed Jun 29, 2020.
[9] Rsipvision.com. Why are we an amazing partner for orthopedic imaging?. URL: https://www.rsipvision.com/orthopedics/. Accessed Jun 29, 2020.
[10] Aicure.com. URL: https://aicure.com/aicure-industry-first-computer-vision-to-capture-dbm/. Accessed Jun 29, 2020.