Breaking News
May 26, 2018 - FDA authorizes marketing of OsteoDetect software for detecting wrist fractures
May 26, 2018 - Children and adolescents growing up in extreme societal conditions more likely to resort to violence
May 26, 2018 - New study puts forth most comprehensive tree of life for malaria parasites
May 26, 2018 - New research shows why babies need to move in the womb
May 26, 2018 - UK steps forward to tackle global antimicrobial resistance
May 26, 2018 - CRISPR-Cas9-based strategy allows researchers to precisely alter hundreds of different genes
May 26, 2018 - UT Southwestern-led researchers find new way to determine prognosis of invasive kidney cancer
May 26, 2018 - Researchers develop film to prevent bacteria from growing on dental retainers and aligners
May 26, 2018 - Mobile health intervention for people with serious mental illness as effective as clinic-based treatment
May 26, 2018 - Vaginal estradiol tablets outperform moisturizers when treating vulvovaginal problems
May 26, 2018 - Researchers call for new genetic tests for congenital diseases
May 26, 2018 - KHN’s ‘What the Health?’ Campaign promises kept, plus ‘nerd reports’
May 26, 2018 - Lung-on-a-chip technology could streamline drug-testing for pulmonary fibrosis
May 26, 2018 - Study finds early antibiotic initiation for majority of premature infants
May 26, 2018 - New environmental monitoring project finds increased numbers of deer ticks in Southern Indiana
May 26, 2018 - Pediatricians Should Advocate for Life Support Training
May 26, 2018 - Cannabidiol significantly reduces seizures in patients with severe form of epilepsy
May 26, 2018 - Allergies can have serious, far-reaching consequences on adolescents
May 26, 2018 - Scientists develop lab-based system to study mechanisms of common liver disease
May 25, 2018 - New guidelines may help pathologists to more accurately classify and diagnose invasive melanoma
May 25, 2018 - Immune cells promote lung cancer metastases by forming clots in tumors, study finds
May 25, 2018 - Can Excess Weight in Toddlers Cause Brain Drain?
May 25, 2018 - Studying insight
May 25, 2018 - Researchers reveal potent new mechanism of action for treatment of IBD
May 25, 2018 - Study shows lack of follow-up care for patients with concussion
May 25, 2018 - Study establishes the importance of haploid cells
May 25, 2018 - Coveted BMJ award bestowed on The Clatterbridge Cancer Center
May 25, 2018 - AACN outlines evidence-based protocols and clinical strategies to manage alarms
May 25, 2018 - Origami inspires researchers to develop new solution for tissue regeneration
May 25, 2018 - Melorheostosis – Genetics Home Reference
May 25, 2018 - Non-addictive pain medication changing therapy for substance use disorders
May 25, 2018 - Delayed lactate measurements in sepsis patients increase risk of in-hospital death
May 25, 2018 - Researchers identify novel epigenetic mutations as cause of neurodevelopmental, congenital disorders
May 25, 2018 - UD researchers examine connection between DNA replication in HPV and cancer
May 25, 2018 - Researchers identify neurons that play key role in aggressive behavior
May 25, 2018 - Snail’s eye inspires new type of RIOCATH urinary catheter
May 25, 2018 - Russian researchers develop high-tech device-transformer for ultrasound examination
May 25, 2018 - Researchers discover unexpected chemosensor pathway for predator odor-evoked innate fear behaviors
May 25, 2018 - Researchers build 3-D printer that offers sweet solution to making detailed structures
May 25, 2018 - Nearly one in three people know someone addicted to opioids
May 25, 2018 - Research suggests link between faulty gene, alcohol, and heart failure
May 25, 2018 - New findings could help fine-tune treatment for cancer patients
May 25, 2018 - New cancer treatment approach targets specific sugar receptors
May 25, 2018 - Skin responsible for uptake of cancer-causing compounds during barbecuing than lungs
May 25, 2018 - Early-onset cannabis use linked to further drug abuse problems
May 25, 2018 - Covered California takes aim at hospital C-section rates
May 25, 2018 - FDA Approves Palynziq (pegvaliase-pqpz) for the Treatment of Adults with Phenylketonuria
May 25, 2018 - Arthritis Glossary
May 25, 2018 - Study links breast cancer to the body’s internal clock
May 25, 2018 - Strenuous exercise in teenage years may protect against height loss later in life
May 25, 2018 - FDA approves novel enzyme therapy for adults with rare and serious genetic disease
May 25, 2018 - New research project aims at developing effective interventions for kids with DLD
May 25, 2018 - Middlemen who save $$ on medicines — but maybe not for you
May 25, 2018 - Study sheds new light on sharp rise in fatal drug overdoses in recent years
May 25, 2018 - Students propose revision of listeriosis guidelines for safer pregnancy
May 25, 2018 - TNFi Exposure In Utero Does Not Up Serious Infection Risk
May 25, 2018 - Organization of cells in the inner ear enables the sense and sensitivity of hearing
May 25, 2018 - Yoga May Be Right Move Against Urinary Incontinence
May 25, 2018 - Drinking recommended amount of milk could protect obese children against metabolic syndrome
May 25, 2018 - New cytokine network can repair tissue damage in the intestine, study finds
May 25, 2018 - Lyme disease researcher dispels misconceptions about ticks and provides prevention tips
May 25, 2018 - Penn researchers find link between social media usage and underage drinking
May 25, 2018 - Unique nanotechnology method to simplify skin disease diagnosis
May 25, 2018 - Study reveals new protective mechanism for tumor cells in breast cancer
May 25, 2018 - FRAME Alternatives Laboratory chosen for major European liver research collaboration
May 25, 2018 - Study shows yogurt may dampen chronic inflammation linked to multiple diseases
May 25, 2018 - Invasive cancers that are born to be bad show detectable differences from harmless tumors
May 25, 2018 - Study identifies new mechanism involved in development of Lou Gehrig’s disease
May 25, 2018 - UAB professor receives award for malaria prevention study in pregnant women in Cameroon
May 25, 2018 - Study provides blueprint of how fruit flies can be used to screen potentially pathogenic human genes
May 25, 2018 - New drug-delivering nanoparticle could offer better way to treat brain tumors
May 25, 2018 - Kessler Foundation scientists compare two tests for assessing learning in individuals with MS
May 25, 2018 - Stroke Symptoms and Diagnosis (Beyond the Basics)
May 25, 2018 - Protein goes against the family to prevent cancer
May 25, 2018 - Drugmakers blamed for blocking generics have milked prices and cost U.S. billions
May 25, 2018 - Speakers announced for National Medicines Symposium 2018
May 25, 2018 - GSK Receives FDA Approval of Arnuity Ellipta for Asthma in Children From 5 Years of Age
May 25, 2018 - Pfizer settles kickback case related to copay assistance for $24m
May 25, 2018 - Nuclear pore functions are essential for T cell survival
May 25, 2018 - Study defines molecular basis to explain connection between mother’s nutrition and infant growth
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