Breaking News
May 21, 2018 - FDA Approves Aimovig (erenumab-aooe) as a Preventive Treatment for Migraine
May 21, 2018 - Infant death study reveals dangerous sleep practices among babysitters, relatives, others
May 21, 2018 - Intermittent fasting may increase diabetes risk, shows study
May 21, 2018 - New study shows declines in prostate cancer screening, diagnoses and treatment
May 21, 2018 - Fasting diets could raise risk of diabetes say experts
May 21, 2018 - FDA Alert: 7K and Poseidon 4500 by Shoreside Enterprises: Voluntary Recall
May 21, 2018 - Cell phones at summer camp: Research explores the effects
May 21, 2018 - Birth rate decline driven by waiting longer to have children, cost of infertility treatment
May 21, 2018 - In-hospital opioid prescribing may increase post-discharge opioid use, shows study
May 21, 2018 - ABPI expert urges to find new ‘blockbuster treatments’ for brain tumors
May 21, 2018 - Disruption of Circadian Rhythm Negatively Impacts Mental Health
May 21, 2018 - Researchers reveal mechanisms of periodic paralysis in people with rare genetic disorder
May 21, 2018 - World first use of cognitive training reduces gait freezing in Parkinson’s patients
May 21, 2018 - NIH stops alcohol study that was looking at purported health benefits of drinking
May 21, 2018 - Higher belly fat levels linked to greater risk of vitamin D deficiency
May 21, 2018 - Scientists collate evidence for mismatch between past evolutionary adaptation and modern lives
May 21, 2018 - New case report reveals negative clinical impact of using biotin supplement
May 21, 2018 - Researchers discover new disease mechanism in chronic tobacco smokers
May 21, 2018 - Breast Cancer Patients May Shorten Herceptin Regimen: Study
May 21, 2018 - Bias keeps women with higher body weights away from the doctor: study
May 21, 2018 - Researchers identify protein essential for eye lens development and clear vision
May 21, 2018 - Frontal cortical lesions moderate response to prism adaptation treatment after stroke
May 21, 2018 - Ultrasound guidelines can reliably differentiate between pediatric thyroid nodules that require biopsy
May 21, 2018 - Weight loss is an important predictor of cancer
May 21, 2018 - Ozone exposure at birth linked to increased risk of developing asthma in childhood
May 21, 2018 - CT scan still effective to determine thrombectomy treatment in stroke, study shows
May 21, 2018 - Clot busting drug combo reduces risk of major strokes in high risk patients
May 21, 2018 - New airway transplantation technique shows promising results in lung cancer patients
May 21, 2018 - Biomarker blood test does not appear to curb antibiotic overuse, shows new study
May 21, 2018 - Lilly’s Galcanezumab Meets Primary Endpoint in Phase 3 Study Evaluating Galcanezumab for the Prevention of Episodic Cluster Headache
May 21, 2018 - Grief symptoms similar in donor vs non-donor decision families
May 21, 2018 - Congo to start vaccinating populations against Ebola today to combat outbreak
May 21, 2018 - Researchers use MR spectroscopy to investigate mechanisms behind targeted treatment for gliomas
May 21, 2018 - Study reveals why older workers have higher stress levels than younger colleagues
May 21, 2018 - Health Tip: Taming a Pollen Allergy
May 21, 2018 - Inducing labor at 39 weeks reduces risks of C-section and other complications
May 20, 2018 - Developmental psychotherapy aims at helping antisocial adolescents become responsible adults
May 20, 2018 - People with OCD process emotions differently than their unaffected siblings
May 20, 2018 - Interfering with enzyme’s movement may be new approach for developing of anti-cancer drugs
May 20, 2018 - Prestroke and poststroke oral anticoagulation therapy in AF patients
May 20, 2018 - Why drug users prefer heroin at home, but cocaine while out
May 20, 2018 - Gene therapy that reverses blindness in dogs could also help treat humans
May 20, 2018 - Opioid-Related Payments Linked to Increase in Opioid Rx
May 20, 2018 - Phone apps push people to take their pills
May 20, 2018 - Backbreaking Work May Shorten Men’s Lives
May 20, 2018 - Harsher drug laws won’t stop violence, argues former police chief
May 20, 2018 - Cognitive decline in dementia is not reduced by exercise
May 20, 2018 - Detecting breast cancer with non-invasive ‘disease screening pill’
May 20, 2018 - Simple treatment may minimize hearing loss triggered by loud noises
May 20, 2018 - Alignment of mother and offspring body clock could prevent diseases such as heart disease and obesity
May 20, 2018 - New commercial data warehouse for life sciences
May 20, 2018 - Practice Intervention Targeting IV Opioids May Cut Exposure
May 20, 2018 - New study provides insight into blood signatures of inflammation
May 20, 2018 - Scientists make breakthrough discovery about vitamin B12
May 20, 2018 - What Causes Cancer? Misconceptions Abound
May 20, 2018 - Proper burial of dead cells limits inflammation
May 20, 2018 - Study offers novel solution to suppress metastatic spread of deadly breast cancer
May 20, 2018 - Perspectives of patients and caregivers on care transitions
May 20, 2018 - Guidelines for weight gain in pregnancy should be changed for underweight and very obese women
May 20, 2018 - Researchers transplant retinal sheets derived from human embryonic stem cells in retinal degeneration mouse models
May 20, 2018 - U.S. military personnel at greater risk for skin cancer than general population
May 20, 2018 - Your immune system holds the line against repeat invaders, thanks to this molecule
May 20, 2018 - Between death and deportation
May 20, 2018 - Developing a High Throughput Mass Spectrometry Platform for Drug Discovery
May 19, 2018 - New project aims to increase awareness among hospital clinicians of non-beneficial treatment at end-of-life
May 19, 2018 - Automated bone scan index offers accurate, speedy prognostic information about prostate cancer
May 19, 2018 - Rutgers Cancer Institute nurses research various topics to enhance patient experience
May 19, 2018 - Computer models provide valuable insight to structure and function of Ebola, Zika viruses
May 19, 2018 - Study exposes key tactic used by deadly fungus
May 19, 2018 - Bacterial signals are crucial to development of pre-leukemic myeloproliferation, study shows
May 19, 2018 - Global experts identify key issues in supporting children with brain injuries transition back to school
May 19, 2018 - Social connections may protect black men who have sex with men from acquiring HIV
May 19, 2018 - Study IDs Factors Linked to Quality of Life With Dementia
May 19, 2018 - Potassium — Consumer
May 19, 2018 - HIV-1 viruses transmitted at birth are resistant to antibodies in mother’s blood
May 19, 2018 - Some water pitchers are much better at removing toxins, shows research
May 19, 2018 - Scientists discover how unusually long strands of RNA help colon cancer cells avoid death
May 19, 2018 - International study finds viable treatment option for people with mild asthma
May 19, 2018 - Mayo discovery could enable development of personalized ovarian, brain cancer treatments
May 19, 2018 - ‘Superbug’ Surfaces at Poultry Farm in China
The Artificial Brain as Doctor

The Artificial Brain as Doctor

image_pdfDownload PDFimage_print

Computers are learning to distinguish whether skin lesion images are benign or malignant and require further treatment. It appears they are on par with, if not more accurate than, many dermatologists, a study shows.

Stanford University researchers compared a computer-driven deep neural network with dermatologists’ ability to visually classify possible melanoma and basal cell or squamous cell carcinoma lesions.

“We recruited 21 board-certified dermatologists and showed them 300 images, where they had to classify the lesions as benign or malignant, and whether they would biopsy or reassure the patient,” said the study’s lead author, Roberto A. Novoa, MD, of Stanford University in California. “From there, the algorithm performed about as well as the dermatologists, if not better. There were dermatologists who did perform better than the algorithm but, in general, this was a proof of concept study, so … we were demonstrating the efficacy of these algorithms for making the diagnosis.”

Computers That Think Like Humans

Deep neural networks are a type of computer algorithm within the field of artificial intelligence, which uses computers to mimic brain function, including reasoning.

While it might sound like a new concept, it’s not. Researchers reported on the potential for computers to diagnose facial tumors in 1986.

“The idea for neural networks has been around since the 1960s,” Novoa said. “But it was only in the last 5 years that the computing power and technological capabilities caught up to the math.”

Today’s neural networks take in vast amounts of information. They then learn the rules that lie behind the data to derive patterns and, eventually, correct answers, Novoa explained.

“Essentially, deep neural networks are mathematical equations that are stacked in layers. They start at the most basic level by learning the edges of all the objects in an image. Then, they move on to telling you this is a triangle or a square. The next layer might indicate whether an image is a cat or a dog. Finally, there’s a layer that says this is a Belgian Malinois or a German Shepherd … If it gets the answer wrong, it goes back through the equation and changes the values and weights of that equation until it gets the most answers correct for the most number of images. By doing so, it learns, over time, what’s important and what’s not.”

From Dogs to Skin Cancer

The genesis of the research, published in Nature last year, began 3 years ago, when Novoa said he saw the advances in the field of deep learning, where these computer programs could differentiate on their own between Belgian Shepherds and German Shepherds.

“I thought if we can do this for dogs, we can do this for skin cancer,” he said.

Novoa and his colleagues at Stanford gathered a dataset of nearly 130,000 images from the internet, including open source images and images from Stanford databases. The researchers went through the data and created a visual taxonomy for the more than 2,000 disease categories, whittling those down to 10 different general categories. The team then used the dataset in the deep neural network’s classifier.

The researchers measured the algorithm’s performance by creating a sensitivity-specificity curve. Sensitivity represented the algorithm’s ability to correctly identify malignant lesions, and specificity represented its ability to accurately identify benign lesions. Assessing the algorithm through the diagnostic tasks of keratinocyte carcinoma classification, melanoma classification, and melanoma classification when viewed using dermoscopy, the researchers found that the algorithm matched the dermatologists’ performance in all three tasks, with the area under the sensitivity-specificity curve amounting to at least 91% of the total area of the graph, according to a Stanford news release.

The deep neural network, in this case, had been trained to classify skin cancer, but the study was done on images of skin cancer that the network had not yet seen. “If you test it on images it has already seen, it already knows the answer,” Novoa said.

The images in the study already had biopsy-proven results. This allowed the researchers to know the true answers when testing the deep neural network and dermatologists.

The next step is to test the algorithm in the real world, Novoa said. “We’re currently putting together a clinical trial to see how well it will perform in the real world. We want to see how well it performs with real patients, with a smartphone camera and a variety of lighting.”

There have been other studies using artificial intelligence and deep neural networks to diagnose melanoma, but this one is different in a few ways, Novoa explained. “Most of the other studies were performed using dermoscopic images. Dermoscopy kind of limits the number of variables. It’s always taken from the same distance and with pretty similar lighting.

“This was looking at clinical images, but they were from a variety of different distances and angles. This also was looking at melanoma, squamous cell carcinoma, and basal cell carcinomas. So, there was a wider group of images. We used a pretty large dataset — it’s among the larger datasets we’ve seen to date.”

Implications

The basic idea of neural networks is that they do pattern recognition, just as a dermatologist would learn patterns and visual data, Novoa said.

Deep neural networks and artificial intelligence may have a growing role in dermatology to help dermatologists provide better care, but not to replace the need for human expertise. While the computer algorithms might be used to triage or screen patients, ultimately a healthcare professional will take responsibility (and liability) for the care.

Algorithms, though, still need improvement.

“If there are biases in the dataset, these can be introduced into the algorithm results,” he continued. “So, for a long time there is going to be a need for supervision of these results in order to have things work optimally. For example, if the algorithm has a ruler in the image, it’s more likely to call it cancer. Why? Because on average, images in our dataset that have rulers are more likely to be malignant. That’s just a small example of the kinds of biases that can be introduced into the data.”

The near future of the technology has far-reaching implications, including the potential development of such things as a smartphone-compatible algorithm, where consumers could capture a lesion in an image and the phone would screen it for skin cancer probability.

“Technology has been changing medicine and all of human endeavors for hundreds of years, but it hasn’t eliminated the need for doctors,” Novoa said. “I don’t think we’re going anywhere. I think dermatologists will be able to adapt and take advantage of these technologies to take care of patients.”

This article originally appeared on the website of our partner Dermatology Times, which is a part of UBM Medica. (Free registration is required.)

2018-01-15T13:00:00-0500

Tagged with:

About author

Related Articles