Abstract
.Significance: Melanoma is a deadly cancer that physicians struggle to diagnose early because they lack the knowledge to differentiate benign from malignant lesions. Deep machine learning approaches to image analysis offer promise but lack the transparency to be widely adopted as stand-alone diagnostics.Aim: We aimed to create a transparent machine learning technology (i.e., not deep learning) to discriminate melanomas from nevi in dermoscopy images and an interface for sensory cue integration.Approach: Imaging biomarker cues (IBCs) fed ensemble machine learning classifier (Eclass) training while raw images fed deep learning classifier training. We compared the areas under the diagnostic receiver operator curves.Results: Our interpretable machine learning algorithm outperformed the leading deep-learning approach 75% of the time. The user interface displayed only the diagnostic imaging biomarkers as IBCs.Conclusions: From a translational perspective, Eclass is better than convolutional machine learning diagnosis in that physicians can embrace it faster than black box outputs. Imaging biomarkers cues may be used during sensory cue integration in clinical screening. Our method may be applied to other image-based diagnostic analyses, including pathology and radiology.
Highlights
Melanoma is the most dangerous skin cancer and the leading cause of death from skin disease
ensemble machine learning classifier (Eclass) trained several independent machine learning algorithms 1000 times in 150 s compared to the Convolutional neural networks (CNN) model, which trained 10 times in 52 h on an Nvidia Quadro M5000 GPU
In a Monte Carlo simulation that randomly drew receiver operator characteristic (ROC) from the 10 CNN ROCs to compare to ROCs randomly drawn from the 1000 Eclass ROCs, the area under the receiver operator characteristic curve (AUROC) was greater for Eclass 74.88% of the time
Summary
Melanoma is the most dangerous skin cancer and the leading cause of death from skin disease. There are over 96,000 new cases in the USA annually with nearly 10,000 deaths attributed to melanoma. Gaps exist in providing specialty care for patients with suspicious lesions and for increasing the numbers of patients seeking potentially life-saving melanoma diagnosis from primary care providers. Despite evidence that early detection increases survival, and despite the need for technology to enhance screening to an expert level on a wide scale, there is uncertainty regarding the effectiveness of state-of-the-art technological methodology in clinical dermatologist screening.[1] Improved screening may prevent melanoma deaths (7230 in the USA in 20192) while decreasing unnecessary invasive procedures because screening guides the binary decision for or against surgical biopsy. Because the skin is accessible to relatively inexpensive and noninvasive diagnostic imaging, and because clinicians rely heavily on visual inspection, melanoma is an ideal case study
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
