A survey on deep learning in medical image analysis

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.

/pdf/a-survey-on-image-data-augmentation-for-deep-learning-4ru1bj2hbb.pdf

A survey on Image Data Augmentation for Deep Learning

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.

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

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.

/pdf/gender-shades-intersectional-accuracy-disparities-in-4qgeu0c1i3.pdf

Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification

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.

Efficient Processing of Deep Neural Networks: A Tutorial and Survey

AD: atopic dermatitis JAK: Janus kinase IL: interleukin INTRODUCTION The management of moderate-to-severe atopic dermatitis (AD) can be challenging in patients refractory to first-line systemic therapies, and inadequate bridging when transitioning between systemic therapies can trigger disease flares. Simultaneously, the ongoing COVID-19 pandemic raises a concern for iatrogenic immunosuppression, particularly with conventional agents, including corticosteroids, cyclosporine, methotrexate, and mycophenolate mofetil. Newer agents may offer less immunosuppression. Dupilumab, a targeted interleukin (IL) 4Ra monoclonal antibody, is Food and Drug Administration-approved for moderate-to-severe AD and has revolutionized AD treatment. However, only 40% of patients achieve clear or almost clear skin after 12 weeks of monotherapy, highlighting the need for additional treatment options. The Janus kinase (JAK)-signal transducers and activators of transcription pathway is involved in downstream signaling of many proinflammatory cytokines, including those within the Th2 axis (e.g., IL-4, IL-5, and IL-13), and oral JAK inhibitors have emerged as a promising therapeutic class for AD. Several recent clinical trials have investigated selective JAK1 inhibitors as monotherapy for moderateto-severe AD, including recent promising results from a phase III study of abrocitinib and a phase II study of upadacitinib. However, in clinical practice,

Bridging to a selective Janus kinase 1 inhibitor in severe atopic dermatitis: An instructive case with upadacitinib.

Leukonychia refers to a white discoloration of the nails. Although several conditions may cause white nails, a rare, isolated, congenital form of the disease is hypothesized to stem from disordered keratinization of the nail plate. Herein, we report a case of a 41-year-old woman with congenital leukonychia and review prior cases.

/pdf/a-case-and-review-of-congenital-leukonychia-484k0o6thb.pdf

A case and review of congenital leukonychia

This cross-sectional survey finds that most dermatologists are receptive to assistance from AI tools and envision a role for augmented intelligence in clinical practice, but they also value the physician-patient relationship and human accuracy

Dermatologists' Perspectives on Artificial Intelligence and Augmented Intelligence - A Cross-sectional Survey.

Alopecia areata (AA) is an autoimmune disease characterized by nonscarring hair loss. As a clinically heterogeneous disease, various classification systems have evolved for defining its severity. In this high-level review of the literature, we discuss the traditional classification systems for AA severity and their strengths and weaknesses. Most recent classifications have focused on the extent of scalp hair loss as a defining feature, but additional clinical aspects of the disease, including location, pattern, and duration of hair loss as well as impact on the patient's quality of life, are also relevant. These various components have typically been used unidimensionally to classify patients. We propose a multidimensional framework to define AA severity that incorporates multiple patient- and illness-related domains. Using such a framework, dermatologists may better assess the severity of the disease for the individual patient beyond the extent of hair loss. 

https://link.springer.com/content/pdf/10.1007/s13555-022-00711-3.pdf

Defining Severity in Alopecia Areata: Current Perspectives and a Multidimensional Framework

This study is the first double-blinded, randomized comparison of two absorbable sutures. To better understand product characteristics and surgeon preference, we conducted a study of two similar-appearing FDA-approved sutures, glyconate and poliglecaprone 25. Four dermatologic surgeons were enlisted. A total of 48 patients with 53 surgical sites were examined. One half of each surgical wound was closed with one type of suture and the other half with the other type. Each half was evaluated for product characteristics. There was no statistically significant difference in surgeon preference for glyconate versus poliglecaprone 25 (P=0.64). Of the cohort preferring poliglecaprone 25, there was a correlation with speed of closure (P=0.06). Of the surgeons that preferred glyconate, we found significantly better visibility (P=0.03), reduced suture breakage during knot tying (P=0.05), and correlation with better handling properties (P=0.06) associated with that preference. The data from this study will enable products to be designed towards these needs and allow surgeons to select sutures that more precisely fit their particular requirements.

Justin M. Ko

Papers

Bridging to a selective Janus kinase 1 inhibitor in severe atopic dermatitis: An instructive case with upadacitinib.

A case and review of congenital leukonychia

Dermatologists' Perspectives on Artificial Intelligence and Augmented Intelligence - A Cross-sectional Survey.

Defining Severity in Alopecia Areata: Current Perspectives and a Multidimensional Framework

A randomized, prospective trial evaluating surgeon preference in selection of absorbable suture material.