Artificial Intelligence in Dermatology: A Primer.

Artificial intelligence is becoming increasingly important in dermatology, with studies reporting accuracy matching or exceeding dermatologists for the diagnosis of skin lesions from clinical and dermoscopic images. However, real-world clinical validation is currently lacking. We review dermatological applications of deep learning, the leading artificial intelligence technology for image analysis, and discuss its current capabilities, potential failure modes, and challenges surrounding performance assessment and interpretability. We address the following three primary applications: (i) teledermatology, including triage for referral to dermatologists; (ii) augmenting clinical assessment during face-to-face visits; and (iii) dermatopathology. We discuss equity and ethical issues related to future clinical adoption and recommend specific standardization of metrics for reporting model performance.

Obscuring surface anatomy in volumetric imaging data.

The identifying or sensitive anatomical features in MR and CT images used in research raise patient privacy concerns when such data are shared. In order to protect human subject privacy, we developed a method of anatomical surface modification and investigated the effects of such modification on image statistics and common neuroimaging processing tools. Common approaches to obscuring facial features typically remove large portions of the voxels. The approach described here focuses on blurring the anatomical surface instead, to avoid impinging on areas of interest and hard edges that can confuse processing tools. The algorithm proceeds by extracting a thin boundary layer containing surface anatomy from a region of interest. This layer is then "stretched" and "flattened" to fit into a thin "box" volume. After smoothing along a plane roughly parallel to anatomy surface, this volume is transformed back onto the boundary layer of the original data. The above method, named normalized anterior filtering, was coded in MATLAB and applied on a number of high resolution MR and CT scans. To test its effect on automated tools, we compared the output of selected common skull stripping and MR gain field correction methods used on unmodified and obscured data. With this paper, we hope to improve the understanding of the effect of surface deformation approaches on the quality of de-identified data and to provide a useful de-identification tool for MR and CT acquisitions.

Evidence of photo manipulation in a delusional parasitosis paper.

In 2004, an article in the Journal of the New York Entomological Society claimed that individuals with delusory parasitosis actually suffer from collembola infestations. The article has been critiqued for poor methodology and results that contradict all knowledge about collembolans. Less easily accounted for has been a figure in the article purporting to show a collembolan in a skin scraping. The image appears to have been altered using photo manipulation software to an unacceptable degree, and this paper demonstrates that to be the case. The altered figure represents creation of nonexistent data, a form of scientific misconduct. Whether the deception is deliberate or a product of an otherwise well-meaning author ignorant of the limits of acceptable image manipulation is unknown, but the result is peer-reviewed support for a conclusion that complicates patient treatment. In the current era of computers, even regional entomology journals must have detailed standards for what kind of images and image manipulations are acceptable for publication.

A physicist at Woods Hole: introducing the image intensifier and image processing to cell biology.

In 1963, by invitation, particle physicist George Reynolds (Princeton University) brought an image intensifier to Woods Hole Marine Biological Laboratory. Together, he and a group of biologists began experimenting with the device as a way to create images of cells in low-light level situations, especially in the study of bioluminescence. In this paper I am interested in how the scientists, a physicist and biologists in collaboration, assessed, interpreted and presented the pictures that they created with the aid of image intensification. In particular, I consider the problem of 'noise' in the image. The paper ends with an example of how Reynolds and a biologist at Woods Hole contended with the presence of noise in images used for publication. Here is an example of how data is modified, that is, enhanced, to serve as scientific evidence. By presenting an early and simple case of the altered image I reveal one way scientists addressed the potentiality of presenting inappropriately modified data - a concern that has garnered much attention in the current age of digital imaging technologies.

Generalizing on best practices in image processing: a model for promoting research integrity: Commentary on: Avoiding twisted pixels: ethical guidelines for the appropriate use and manipulation of scientific digital images.

Modifying images for scientific publication is now quick and easy due to changes in technology. This has created a need for new image processing guidelines and attitudes, such as those offered to the research community by Doug Cromey (Cromey 2010). We suggest that related changes in technology have simplified the task of detecting misconduct for journal editors as well as researchers, and that this simplification has caused a shift in the responsibility for reporting misconduct. We also argue that the concept of best practices in image processing can serve as a general model for education in best practices in research.

Avoiding twisted pixels: ethical guidelines for the appropriate use and manipulation of scientific digital images.

Digital imaging has provided scientists with new opportunities to acquire and manipulate data using techniques that were difficult or impossible to employ in the past. Because digital images are easier to manipulate than film images, new problems have emerged. One growing concern in the scientific community is that digital images are not being handled with sufficient care. The problem is twofold: (1) the very small, yet troubling, number of intentional falsifications that have been identified, and (2) the more common unintentional, inappropriate manipulation of images for publication. Journals and professional societies have begun to address the issue with specific digital imaging guidelines. Unfortunately, the guidelines provided often do not come with instructions to explain their importance. Thus they deal with what should or should not be done, but not the associated 'why' that is required for understanding the rules. This article proposes 12 guidelines for scientific digital image manipulation and discusses the technical reasons behind these guidelines. These guidelines can be incorporated into lab meetings and graduate student training in order to provoke discussion and begin to bring an end to the culture of "data beautification".

Against abuse of digital photography techniques in morphology--Ethical Code of Slovak Anatomical Society.

Digital graphical methods allow extensive manipulation of pictures literally beyond any limits. By such methods, it is possible to change original information obtained by morphological observation, even to fraudulently produce distorted results or forgeries. For this reason, basic rules were proposed defining what is allowed and what is not accepted during picture processing. These rules were discussed and approved by a plenary meeting of Slovak Anatomical Society on September 9th, 2007 under the name Ethical Code of Slovak Anatomical Society. We call on all potential authors of publications and dissertation works to obey the rules of Ethical Code of Slovak Anatomical Society and thus to prevent any doubts which may arise about the faithfulness of published materials in morphological or other disciplines, which use a picture as an evidence or illustration means (Fig. 3). Full Text (Free, PDF) www.bmj.sk.

The digital revolution.

Digital image technology has advanced considerably in recent years and it has never been simpler to capture digital images or send them around the world. This new technology can bring tremendous benefits to healthcare but, if used inappropriately, will compromise patient confidentiality and increase the risk of litigation to the health providers. It is now time to assess the risks and agree safeguards which will promote the safe use of digital technology in the healthcare environment.