Breaking News
January 18, 2019 - Let’s map our DNA and save billions each year in health costs
January 18, 2019 - AI demonstrates potential to identify irregular heart rhythms as well as humans
January 17, 2019 - Study shows link between air pollution and increased risk of sleep apnea
January 17, 2019 - Neck-strengthening exercises can protect athletes from concussions
January 17, 2019 - Computer model shows how to better control MRSA outbreaks
January 17, 2019 - Pain is unpleasant, and now scientists have identified the set of responsible neurons
January 17, 2019 - CUIMC Celebrates 2018-2019
January 17, 2019 - Study reveals potential pathway for endothelial cells to avoid apoptosis
January 17, 2019 - Hamilton Storage launches LabElite DeCapper SL to expand LabElite product family
January 17, 2019 - Location of epigenetic changes co-locate with genetic signal causing psychartric disorder
January 17, 2019 - Researchers awarded 6.1 million euros to address female fertility problems
January 17, 2019 - Counseling appointments fail to reduce weight gain during pregnancy, shows study
January 17, 2019 - Contraceptive patch that could provide 6 months of contraception within seconds
January 17, 2019 - Yeast model may pave way for development of novel therapies for metabolic disorders
January 17, 2019 - Study determines impact of antibiotic perturbation of the gut microbiome on skeletal health
January 17, 2019 - Cardiometabolic Risk Up With Tourette, Chronic Tic Disorder
January 17, 2019 - Hong Kong scientists claim ‘broad-spectrum’ antiviral breakthrough
January 17, 2019 - Researchers discover the brain cells that make pain unpleasant | News Center
January 17, 2019 - Hepatitis Is Common in New Cancer Patients
January 17, 2019 - Podcast: KHN’s ‘What The Health?’ Drug Prices Are Rising Again. Is Someone Going To Do Something About It?
January 17, 2019 - Smoking significantly increases your biological age, study shows
January 17, 2019 - B-group vitamins may be beneficial for people with first episode psychosis
January 17, 2019 - Researchers demonstrate how manganese produces parkinsonian syndrome
January 17, 2019 - Researchers suggest link between personality type and attitude towards others’ bodies
January 17, 2019 - Mutant mice administered with cocaine failed to exhibit hyperactivity, shows study
January 17, 2019 - Health Tip: Understanding a Heart Murmur
January 17, 2019 - Gut protein mutations shield against spikes in glucose
January 17, 2019 - Engineered immune cells target broad range of pediatric solid tumors in mice | News Center
January 17, 2019 - Study finds link between high pesticide exposure and poor sense of smell among farmers
January 17, 2019 - Many cancer patients have undiagnosed hepatitis
January 17, 2019 - New study finds only 13% of outpatient antibiotic prescriptions to be appropriate
January 17, 2019 - Stem cell-based approach to diabetes offers hope for treatment
January 17, 2019 - New project receives €8.65 million from EU and Canada to ease genomic, health data sharing
January 17, 2019 - Improvements in pharmacological study to fight cognitive impairment in schizophrenia
January 17, 2019 - Study looks at trends over time in oral antibiotic prescribing by dermatologists
January 17, 2019 - Most substance use disorder treatment facilities do not offer medication treatment
January 17, 2019 - Multiple sclerosis could benefit from stem cell therapy
January 17, 2019 - Researchers manipulate T cells to improve transplant success
January 17, 2019 - Put away your rulers and reach for your phone
January 17, 2019 - Mindfulness linked with fewer menopausal symptoms
January 17, 2019 - Integrated care to women with PMADs offered at several levels
January 17, 2019 - Researchers identify MANF as a rejuvenating factor in parabiosis
January 17, 2019 - Truncal mutations study suggests new direction in origins of cancer
January 17, 2019 - Beckman Coulter launches new ClearLLab 10C System for clinical flow cytometry lab
January 17, 2019 - Effects of linoleic acid on the body are largely dependent on genes, shows study
January 17, 2019 - Pre-injury exercise reduces damage to both muscles and nerves, study finds
January 17, 2019 - Minimizing Antibody Size to Maximize Research Potential
January 17, 2019 - Research finds large genome in tiny forest defoliator
January 17, 2019 - Technology helps reduce the yearning for unhealthy food
January 17, 2019 - Imec develops prototype cardiovascular device
January 17, 2019 - New Drug Application for Insomnia Disorder Treatment Lemborexant Submitted in the United States
January 17, 2019 - What you should know about teeth whitening
January 17, 2019 - Why Older Adults Should Eat More Protein (And Not Overdo Protein Shakes)
January 17, 2019 - Colorectal cancer mortality rates predicted to increase globally
January 17, 2019 - Scientists discover mutational signatures of tumor hypoxia
January 17, 2019 - New evidence shows how fever alters immune cells
January 17, 2019 - Researchers find new class of blood pressure-regulating peptides in vampire bat venom
January 17, 2019 - Promega to exhibit new Maxwell RSC48 platform at 2019 Festival of Genomics
January 17, 2019 - Study pinpoints immune cells that could be key to tackling hypertension
January 17, 2019 - Couples Intervention May Aid Partners of Diabetes Patients
January 17, 2019 - Your weight history may predict your heart failure risk
January 17, 2019 - Explore a cornucopia of accomplishments in prematurity research
January 17, 2019 - New study identifies four characteristics that predict severity of postpartum depression
January 17, 2019 - AHF urges United Nations to follow own mandate for protecting Ebola response efforts
January 17, 2019 - New, scalpel-free treatment for reducing Parkinson’s tremor gets FDA approval
January 17, 2019 - Neurobiologists uncover key component of how the human brain marks time
January 17, 2019 - LifeTime receives fund to develop a plan to embed its vision for healthier future
January 17, 2019 - WTC first responders at higher risk for head and neck cancers, study finds
January 17, 2019 - New NSF funded study may help physicians decrease brain injury deaths
January 17, 2019 - Ham bones contain peptides that could have cardioprotective effects
January 17, 2019 - Research finds how Candida albicans adapt to low oxygen levels to cause infection
January 17, 2019 - Cobra Biologics announces appointment of Dr Darrell Sleep as Director of Innovation
January 17, 2019 - Cellular protein that interacts with viruses appears to enable infection process of Zika virus
January 17, 2019 - Opioids Now More Deadly for Americans Than Traffic Accidents
January 17, 2019 - Women who start periods early are at greater risk of cardiovascular problems
January 17, 2019 - The brain-circuitry clash that keeps you from diving into that plate of ribs when you’re dining with royalty
January 17, 2019 - Poo transplant can successfully treat patients with ulcerative colitis
January 17, 2019 - Study suggests key role for glial cells in Parkinson’s disease
January 17, 2019 - Educational videos in clinical settings increase HPV vaccination rates among adolescents
January 17, 2019 - Better understanding of aggressive brain tumour
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