Justin M. Ko

Bio: Justin M. Ko is an academic researcher from Stanford University. The author has contributed to research in topics: Medicine & Alopecia areata. The author has an hindex of 15, co-authored 62 publications receiving 7124 citations. Previous affiliations of Justin M. Ko include Harvard University & Eastern Virginia Medical School.

Dermatologist-level classification of skin cancer with deep neural networks

Abstract: Skin cancer, the most common human malignancy, is primarily diagnosed visually, beginning with an initial clinical screening and followed potentially by dermoscopic analysis, a biopsy and histopathological examination. Automated classification of skin lesions using images is a challenging task owing to the fine-grained variability in the appearance of skin lesions. Deep convolutional neural networks (CNNs) show potential for general and highly variable tasks across many fine-grained object categories. Here we demonstrate classification of skin lesions using a single CNN, trained end-to-end from images directly, using only pixels and disease labels as inputs. We train a CNN using a dataset of 129,450 clinical images-two orders of magnitude larger than previous datasets-consisting of 2,032 different diseases. We test its performance against 21 board-certified dermatologists on biopsy-proven clinical images with two critical binary classification use cases: keratinocyte carcinomas versus benign seborrheic keratoses; and malignant melanomas versus benign nevi. The first case represents the identification of the most common cancers, the second represents the identification of the deadliest skin cancer. The CNN achieves performance on par with all tested experts across both tasks, demonstrating an artificial intelligence capable of classifying skin cancer with a level of competence comparable to dermatologists. Outfitted with deep neural networks, mobile devices can potentially extend the reach of dermatologists outside of the clinic. It is projected that 6.3 billion smartphone subscriptions will exist by the year 2021 (ref. 13) and can therefore potentially provide low-cost universal access to vital diagnostic care.

Induction and exacerbation of psoriasis with TNF-blockade therapy: A review and analysis of 127 cases

Abstract: Background: There are reports of rare adverse effects of tumor necrosis factor (TNF) inhibitors, including infections, malignancies, and induction of autoimmune conditions. Intriguing, are cases of induction or exacerbation of psoriasis in conjunction with TNF inhibitor therapy, given that they are approved for treatment of the same condition. Objective: Published cases of psoriasis occurring during anti-TNF therapy were analyzed, including overviews of proposed etiologies and treatment recommendations. Methods: A literature search using Ovid MEDLINE and PubMed was performed for articles published between January 1990 and September 2007 to collect reported cases of psoriasis in patients receiving therapy with TNF blocking agents. Results: A total of 127 cases were identified: 70 in patients on infliximab (55.1%), 35 with etanercept (27.6%), and 22 with adalimumab (17.3%). Females comprised 58% of cases; mean age of reported patients was 45.8 years, and the time from initiation of treatment to onset of les...

Safety and efficacy of the JAK inhibitor tofacitinib citrate in patients with alopecia areata

Abstract: BACKGROUND. Alopecia areata (AA) is an autoimmune disease characterized by hair loss mediated by CD8+ T cells. There are no reliably effective therapies for AA. Based on recent developments in the understanding of the pathomechanism of AA, JAK inhibitors appear to be a therapeutic option; however, their efficacy for the treatment of AA has not been systematically examined. METHODS. This was a 2-center, open-label, single-arm trial using the pan-JAK inhibitor, tofacitinib citrate, for AA with >50% scalp hair loss, alopecia totalis (AT), and alopecia universalis (AU). Tofacitinib (5 mg) was given twice daily for 3 months. Endpoints included regrowth of scalp hair, as assessed by the severity of alopecia tool (SALT), duration of hair growth after completion of therapy, and disease transcriptome. RESULTS. Of 66 subjects treated, 32% experienced 50% or greater improvement in SALT score. AA and ophiasis subtypes were more responsive than AT and AU subtypes. Shorter duration of disease and histological peribulbar inflammation on pretreatment scalp biopsies were associated with improvement in SALT score. Drug cessation resulted in disease relapse in 8.5 weeks. Adverse events were limited to grade I and II infections. An AA responsiveness to JAK/STAT inhibitors score was developed to segregate responders and nonresponders, and the previously developed AA disease activity index score tracked response to treatment. CONCLUSIONS. At the dose and duration studied, tofacitinib is a safe and effective treatment for severe AA, though it does not result in a durable response. Transcriptome changes reveal unexpected molecular complexity within the disease. TRIAL REGISTRATION. ClinicalTrials.gov {"type":"clinical-trial","attrs":{"text":"NCT02197455","term_id":"NCT02197455"}}NCT02197455 and {"type":"clinical-trial","attrs":{"text":"NCT02312882","term_id":"NCT02312882"}}NCT02312882. FUNDING. This work was supported by the US Department of Veterans Affairs Office of Research and Development, National Institute of Arthritis and Musculoskeletal and Skin Diseases National Institutes of Health grant R01 AR47223 and U01 AR67173, the National Psoriasis Foundation, the Swedish Society of Medicine, the Fernstrom Foundation, the Locks of Love Foundation, the National Alopecia Areata Foundation, and the Ranjini and Ajay Poddar Resource Fund for Dermatologic Diseases Research.

Corrigendum: Dermatologist-level classification of skin cancer with deep neural networks

Abstract: Nature 542, 115–118 (2017); doi:10.1038/nature21056 In the Acknowledgements section of this Letter, the sentence: “This study was supported by the Baxter Foundation, California Institute for Regenerative Medicine (CIRM) grants TT3-05501 and RB5-07469 and US National Institutes of Health (NIH) grantsAG044815, AG009521, NS089533, AR063963 and AG020961 (H.M.B.)” should have read: “This study was supported by funding from the Baxter Foundation to H.M.B.” Furthermore, the last line of the Acknowledgements section should have read: “In addition, this work was supported by a National Institutes of Health (NIH) National Center for Advancing Translational Science Clinical and Translational Science Award (UL1 TR001085). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.” The original Letter has been corrected online.

Prospective assessment of computer-aided detection in interpretation of screening mammography.

Abstract: OBJECTIVE. The purpose of this study was to prospectively assess the usefulness of computer-aided detection (CAD) in the interpretation of screening mammography and to provide the true sensitivity and specificity of this technique in a clinical setting.SUBJECTS AND METHODS. Over a 26-month period, 5,016 screening mammograms were interpreted without, and subsequently with, the assistance of the iCAD MammoReader detection system. Data collected for actionable findings included dominant feature (calcification, mass, asymmetry, architectural distortion), detection method (radiologist only, CAD only, or both radiologist and CAD), BI-RADS assessment code, associated histopathology for those undergoing biopsy, and tumor stage for malignant lesions. The study population was cross-checked against an independent reference standard to identify false-negative cases.RESULTS. Of the 5,016 cases, the recall rate increased from 12% to 14% with the addition of CAD. Of the 107 (2%) patients who underwent biopsy, 101 (94%) ...

A survey on Image Data Augmentation for Deep Learning

Abstract: Deep convolutional neural networks have performed remarkably well on many Computer Vision tasks. However, these networks are heavily reliant on big data to avoid overfitting. Overfitting refers to the phenomenon when a network learns a function with very high variance such as to perfectly model the training data. Unfortunately, many application domains do not have access to big data, such as medical image analysis. This survey focuses on Data Augmentation, a data-space solution to the problem of limited data. Data Augmentation encompasses a suite of techniques that enhance the size and quality of training datasets such that better Deep Learning models can be built using them. The image augmentation algorithms discussed in this survey include geometric transformations, color space augmentations, kernel filters, mixing images, random erasing, feature space augmentation, adversarial training, generative adversarial networks, neural style transfer, and meta-learning. The application of augmentation methods based on GANs are heavily covered in this survey. In addition to augmentation techniques, this paper will briefly discuss other characteristics of Data Augmentation such as test-time augmentation, resolution impact, final dataset size, and curriculum learning. This survey will present existing methods for Data Augmentation, promising developments, and meta-level decisions for implementing Data Augmentation. Readers will understand how Data Augmentation can improve the performance of their models and expand limited datasets to take advantage of the capabilities of big data.

High-performance medicine: the convergence of human and artificial intelligence

Abstract: The use of artificial intelligence, and the deep-learning subtype in particular, has been enabled by the use of labeled big data, along with markedly enhanced computing power and cloud storage, across all sectors. In medicine, this is beginning to have an impact at three levels: for clinicians, predominantly via rapid, accurate image interpretation; for health systems, by improving workflow and the potential for reducing medical errors; and for patients, by enabling them to process their own data to promote health. The current limitations, including bias, privacy and security, and lack of transparency, along with the future directions of these applications will be discussed in this article. Over time, marked improvements in accuracy, productivity, and workflow will likely be actualized, but whether that will be used to improve the patient-doctor relationship or facilitate its erosion remains to be seen.

Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification

Abstract: The paper “Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification” by Joy Buolamwini and Timnit Gebru, that will be presented at the Conference on Fairness, Accountability, and Transparency (FAT*) in February 2018, evaluates three commercial API-based classifiers of gender from facial images, including IBM Watson Visual Recognition. The study finds these services to have recognition capabilities that are not balanced over genders and skin tones [1]. In particular, the authors show that the highest error involves images of dark-skinned women, while the most accurate result is for light-skinned men.

Efficient Processing of Deep Neural Networks: A Tutorial and Survey

Abstract: Deep neural networks (DNNs) are currently widely used for many artificial intelligence (AI) applications including computer vision, speech recognition, and robotics. While DNNs deliver state-of-the-art accuracy on many AI tasks, it comes at the cost of high computational complexity. Accordingly, techniques that enable efficient processing of DNNs to improve energy efficiency and throughput without sacrificing application accuracy or increasing hardware cost are critical to the wide deployment of DNNs in AI systems. This article aims to provide a comprehensive tutorial and survey about the recent advances toward the goal of enabling efficient processing of DNNs. Specifically, it will provide an overview of DNNs, discuss various hardware platforms and architectures that support DNNs, and highlight key trends in reducing the computation cost of DNNs either solely via hardware design changes or via joint hardware design and DNN algorithm changes. It will also summarize various development resources that enable researchers and practitioners to quickly get started in this field, and highlight important benchmarking metrics and design considerations that should be used for evaluating the rapidly growing number of DNN hardware designs, optionally including algorithmic codesigns, being proposed in academia and industry. The reader will take away the following concepts from this article: understand the key design considerations for DNNs; be able to evaluate different DNN hardware implementations with benchmarks and comparison metrics; understand the tradeoffs between various hardware architectures and platforms; be able to evaluate the utility of various DNN design techniques for efficient processing; and understand recent implementation trends and opportunities.