Breaking News
December 17, 2017 - Teens Acting Badly? Smog Could Be to Blame
December 17, 2017 - FDA Says ‘Yes’ to Short-Acting Insulin Admelog
December 17, 2017 - Vaping popular among teens; opioid misuse at historic lows
December 17, 2017 - Lower Urinary Symptoms Occur in Almost All Patients with SSc
December 17, 2017 - Genetic mutation in extended Amish family in Indiana protects against aging and increases longevity (Update)
December 16, 2017 - Butler Hospital launches international Alzheimer’s disease prevention study
December 16, 2017 - iMedicalApps: Virtual Reality Boosts Self-Confidence for Med Students
December 16, 2017 - Researchers validate five new genes responsible for Amyotrophic Lateral Sclerosis
December 16, 2017 - New genetic analysis of candidiasis reveals surprising fungal sex secrets
December 16, 2017 - New high precision machine-learning model could help accelerate drug discovery
December 16, 2017 - Groundbreaking gene therapy trial brings cure for hemophilia closer
December 16, 2017 - Racial Differences Seen in IgG4 Disease
December 16, 2017 - Treacher Collins Syndrome
December 16, 2017 - New approach to tracking how deadly ‘superbugs’ travel could slow their spread
December 16, 2017 - Muscle paralysis may promote breakdown of bones
December 16, 2017 - WSU scientists create injectable dye to track progression of diseases
December 16, 2017 - Kaiser Permanente delivers clot-busting drugs to stroke patients more than twice as fast as national rates
December 16, 2017 - Some Great Holiday Foods for Weight Loss
December 16, 2017 - Shared Decision-Making Strategies for Lung Ca Screening Get High Marks
December 16, 2017 - Lactic acid bacteria can protect against Influenza A virus, study finds
December 16, 2017 - Cancer immunotherapy’s effectiveness may depend on patient’s genetic makeup
December 16, 2017 - Researchers explore patient-doctor conversations, best practices linked to opioid tapering
December 16, 2017 - ‘Virtual child’ to help professionals learn key techniques to treat children with autism
December 16, 2017 - IU scientists discover way to make drug treatment more successful against malaria
December 16, 2017 - Prostate cancer researchers find significant disparities between two liquid biopsy providers
December 16, 2017 - ED-Diagnosed Lung Ca Patients Worse Off: Clin Onc News Report
December 16, 2017 - Calcium in Urine Test: MedlinePlus Lab Test Information
December 16, 2017 - Pregnancy-related conditions taken together leave moms—and dads—at risk
December 16, 2017 - Research uncovers mechanism implicated in defective function of tumor-associated dendritic cells
December 16, 2017 - OncoBreak: Stubborn Racial Disparities; Paid Medical Leave & Chemo; DIY Gene Tests
December 16, 2017 - Critical link between obesity and diabetes has been identified
December 16, 2017 - Transfusion dependence reduces access to high-quality end-of-life care for leukemia patients
December 16, 2017 - Porvair and Suzhou Tianlong Bio to develop epigenetic analysis technologies
December 16, 2017 - FDA Approves Ixifi (infliximab-qbtx), a Biosimilar to Remicade
December 16, 2017 - Morning Break: Trump to Get Check-Up; Cancerous Transplant; Death Knell for MIPS?
December 16, 2017 - First transcatheter implant for diastolic heart failure successful
December 16, 2017 - ‘Sushi-like’ nanodiscs provide structural snapshots of misfolding proteins
December 16, 2017 - Inherited gene variation may be to blame for poor survival of patients with early-onset breast cancer
December 16, 2017 - Sign-up deadline is Friday, but some people may get extra time
December 16, 2017 - Higher Booze Taxes Might Pay Off for Public Health
December 16, 2017 - Regular Activity in Midlife Spares Joints in Women
December 16, 2017 - Rain May Not Cause Achy Joints After All: MedlinePlus Health News
December 16, 2017 - MedDiet adherence doesn’t affect acute heart failure mortality
December 16, 2017 - HKBU experts develop new generation of smart anti-cancer drug molecules
December 16, 2017 - Chronic Kidney Disease Audit finds wide variations in coding of CKD patients in primary care
December 16, 2017 - Scientists use nanoparticles to fight Mucoviscidosis
December 16, 2017 - Increasing physical activity decreases risk of death from lymphoma
December 16, 2017 - Fear compromises the health, well-being of immigrant families, survey finds
December 16, 2017 - Rejected antibiotic candidate could be worth a second look, research finds
December 16, 2017 - Is Nation on the Right Track to Combat Opioid Crisis?
December 16, 2017 - Arthritis No Longer Just a Disease of the Old: MedlinePlus Health News
December 16, 2017 - Study reveals biology behind why muscle stem cells respond differently to aging or injury
December 16, 2017 - Family members without inherited mutation have increased risk of melanoma
December 16, 2017 - Researchers reveal previously unknown mechanism that inhibits cells’ ability to develop into tumors
December 16, 2017 - Studies highlight potential of fMRI applications to detect, treat epilepsy in children
December 16, 2017 - Active surveillance proposed as first-line approach to manage patients with low-risk PMC of the thyroid
December 16, 2017 - Patients’ life values affect their attendance at medical treatment for pelvic-floor dysfunction
December 16, 2017 - Experts consider hazards of antibiotic resistances to be high
December 16, 2017 - Study finds erectile dysfunction as risk factor for early cardiovascular disease
December 16, 2017 - Amber-tinted glasses may reduce insomnia severity
December 16, 2017 - Arthritis Drug Seen Lowering GvHD Risk
December 16, 2017 - Atoh1, a potential Achilles’ heel of Sonic Hedgehog medulloblastoma
December 15, 2017 - Cornell engineers develop new method to measure vital signs using radio waves
December 15, 2017 - Rutgers studies highlight need for salon clients, workers to protect themselves from health risks
December 15, 2017 - FDA Approves Nucala (mepolizumab) for Eosinophilic Granulomatosis with Polyangiitis (Churg-Strauss Syndrome)
December 15, 2017 - Morning Break: CVS Buying Aetna; Uterus Transplant Baby; Your Brain on Drugs, Redux
December 15, 2017 - Social phobia linked to autism and schizophrenia
December 15, 2017 - Timestrip technology helping to prevent missed vaccinations
December 15, 2017 - Researchers win NIH grants for Alzheimer’s research on Amish resilience and rapid onset
December 15, 2017 - Mitochondrial error-correction mechanism essential for energy production of cells
December 15, 2017 - New report reveals steep rise in lung disease admissions to emergency departments during winter
December 15, 2017 - Study finds social stigma as barrier to successful treatment of children with HIV in Ethiopia
December 15, 2017 - Health Tip: Keep Gift-Giving Stress Under Wraps
December 15, 2017 - Long Stoppage of Bisphosphonates Tied to More Fractures
December 15, 2017 - Triglycerides Test: MedlinePlus Lab Test Information
December 15, 2017 - Study shows interventions, though few, can be effective for students with high-functioning autism
December 15, 2017 - Higher blood sugar during first trimester of pregnancy increases child’s risk of congenital heart defect
December 15, 2017 - Study tests accuracy of laboratory-developed cancer tests and FDA-approved companion diagnostics
December 15, 2017 - Extracellular vesicles can be used to effectively delay progression of kidney damage
December 15, 2017 - Targeted lung cancer treatments may benefit smokers and non-smokers alike
Researchers determine ‘sweet spot’ could enhance diagnosis of melanoma skin lesions

Researchers determine ‘sweet spot’ could enhance diagnosis of melanoma skin lesions

image_pdfDownload PDFimage_print

Too much, too little, just right. It might seem like a line from “Goldilocks and the Three Bears,” but actually describes an important finding from researchers in Florida Atlantic University’s College of Engineering and Computer Science. They have developed a technique using machine learning – a sub-field of artificial intelligence (AI) – that will enhance computer-aided diagnosis (CADx) of melanoma. Thanks to the algorithm they created – which can be used in mobile apps that are being developed to diagnose suspicious moles – they were able to determine the “sweet spot” in classifying images of skin lesions.

This new finding, published in the Journal of Digital Imaging, will ultimately help clinicians more reliably identify and diagnose melanoma skin lesions, distinguishing them from other types of skin lesions. The research was conducted in the NSF Industry/University Cooperative Research Center for Advanced Knowledge Enablement (CAKE) at FAU and was funded by the Center’s industry members.

Melanoma is a particularly deadly form of skin cancer when left undiagnosed. In the United States alone, there were an estimated 76,380 new cases of melanoma and an estimated 6,750 deaths due to melanoma in 2016. Malignant melanoma and benign skin lesions often appear very similar to the untrained eye. Over the years, dermatologists have developed different heuristic classification methods to diagnose melanoma, but to limited success (65 to 80 percent accuracy). As a result, computer scientists and doctors have teamed up to develop CADx tools capable of aiding physicians to classify skin lesions, which could potentially save numerous lives each year.

“Contemporary CADx systems are powered by deep learning architectures, which basically consist of a method used to train computers to perform an intelligent task. We feed them massive amounts of data so that they can learn to extract meaning from the data and, eventually, demonstrate performance comparable to human experts – dermatologists in this case,” said Oge Marques, Ph.D., lead author of the study and a professor in FAU’s Department of Computer and Electrical Engineering and Computer Science who specializes in machine cognition, medical imaging and health care technologies. “We are not trying to replace physicians or other medical professionals with artificial intelligence. We are trying to help them solve a problem faster and with greater accuracy, in other words enabling augmented intelligence.”

Images of skin lesions often contain more than just skin lesions – background noise, hair, scars, and other artifacts in the image can potentially confuse the CADx system. To prevent the classifier from incorrectly associating these irrelevant artifacts with melanoma, the images are segmented into two parts, separating the lesion from the surrounding skin, hoping that the segmented lesion can be more easily analyzed and classified.

“Previous studies have produced conflicting results: some research suggests that segmentation improves classification while other research suggests that segmentation is detrimental, due to a loss of contextual information around the lesion area,” said Marques. “How much we segment an image can either help or impede skin lesion classification.”

With that in mind, Marques and his collaborators Borko Furht, Ph.D., a professor in FAU’s Department of Electrical and Computer Engineering and Computer Science and director of the NSF-sponsored CAKE at FAU; Jack Burdick, a second-year master’s student at FAU; and Janet Weinthal, an undergraduate student at FAU, tested their hypothesis: “How much segmentation is too much?”

To test their hypothesis, the researchers first compared the effects of no segmentation, full segmentation, and partial segmentation on classification and demonstrated that partial segmentation led to the best results. They then proceeded to determine how much segmentation would be “just right.” To do that, they used three degrees of partial segmentation, investigating how a variable-sized non-lesion border around the segmented skin lesion affects classification results. They performed comparisons in a systematic and reproducible manner to demonstrate empirically that a certain amount of segmentation border around the lesion could improve classification performance.

Their findings suggest that extending the border beyond the lesion to include a limited amount of background pixels improves their classifier’s ability to distinguish melanoma from a benign skin lesion.

“Our experimental results suggest that there appears to be a ‘sweet spot’ in the degree to which the surrounding skin included is neither too great nor too small and provides a ‘just right’ amount of context,” said Marques.

Their method showed an improvement across all relevant measures of performance for a skin lesion classifier.

Source:

http://www.fau.edu/newsdesk/articles/melanoma-diagnosis.php

Tagged with:

About author

Related Articles