Analysis of Artificial Intelligence-Based Approaches Applied to Non-Invasive Imaging for Early Detection of Melanoma: A Systematic Review.

BACKGROUND: Melanoma, the deadliest form of skin cancer, poses a significant public health challenge worldwide. Early detection is crucial for improved patient outcomes. Non-invasive skin imaging techniques allow for improved diagnostic accuracy; however, their use is often limited due to the need for skilled practitioners trained to interpret images in a standardized fashion. Recent innovations in artificial intelligence (AI)-based techniques for skin lesion image interpretation show potential for the use of AI in the early detection of melanoma. OBJECTIVE: The aim of this study was to evaluate the current state of AI-based techniques used in combination with non-invasive diagnostic imaging modalities including reflectance confocal microscopy (RCM), optical coherence tomography (OCT), and dermoscopy. We also aimed to determine whether the application of AI-based techniques can lead to improved diagnostic accuracy of melanoma. METHODS: A systematic search was conducted via the Medline/PubMed, Cochrane, and Embase databases for eligible publications between 2018 and 2022. Screening methods adhered to the 2020 version of the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Included studies utilized AI-based algorithms for melanoma detection and directly addressed the review objectives. RESULTS: We retrieved 40 papers amongst the three databases. All studies directly comparing the performance of AI-based techniques with dermatologists reported the superior or equivalent performance of AI-based techniques in improving the detection of melanoma. In studies directly comparing algorithm performance on dermoscopy images to dermatologists, AI-based algorithms achieved a higher ROC (>80%) in the detection of melanoma. In these comparative studies using dermoscopic images, the mean algorithm sensitivity was 83.01% and the mean algorithm specificity was 85.58%. Studies evaluating machine learning in conjunction with OCT boasted accuracy of 95%, while studies evaluating RCM reported a mean accuracy rate of 82.72%. CONCLUSIONS: Our results demonstrate the robust potential of AI-based techniques to improve diagnostic accuracy and patient outcomes through the early identification of melanoma. Further studies are needed to assess the generalizability of these AI-based techniques across different populations and skin types, improve standardization in image processing, and further compare the performance of AI-based techniques with board-certified dermatologists to evaluate clinical applicability.

Papuloerythroderma of Ofuji in a Young Man.

Papuloerythroderma of Ofuji (PEO) is a rare skin disorder characterized by a distinctive pattern of pruritic, flat-topped, erythematous papules which coalesce into an erythroderma-like eruption with classic sparing of the skin folds. Although the pathogenesis of this condition is incompletely understood, previous reports have suggested a notable link between PEO and various forms of malignancy and immunocompromised states. Here, we report a case of a healthy young male with no comorbidities who presented with the classical features of PEO that responded well to combination therapy comprised of topical corticosteroids and phototherapy.

Case Report: Severe SARS-CoV-2 Infection Treated with Remdesivir in a Patient with ESRD.

BACKGROUND: The novel coronavirus disease 2019 (COVID-19) has led to a global pandemic since its emergence from Wuhan, China, in December of 2019. As research continues to evolve, there is a paucity of reports describing the management and treatment of COVID-19 in patients with acute kidney failure and End-Stage Renal Disease (ESRD). These patients have increased susceptibility to developing severe clinical symptoms from SARS-CoV-2 infection due to their underlying comorbidities. Remdesivir has emerged as a promising antiviral drug against SARS-CoV-2. However, data regarding the clinical benefits of remdesivir in patients with severe renal impairment is unavailable as they have been excluded from clinical trials due to the risk of sulfobutylether-ß-cyclodextrin (SBECD) accumulation in patients with eGFR<30 ml/min per 1.73m2. CASE PRESENTATION: We present the first case of a 47-year-old male with end-stage renal disease who was successfully treated with remdesivir during hospitalization for acute respiratory distress syndrome and respiratory failure arising from COVID-19. The worsening clinical progress of the patient despite intensive care and treatment with intravenous azithromycin therapy led to the decision to utilize remdesivir after a risk-benefit analysis, despite his eGFR being <15 ml/min per 1.73m2. Although the patient developed reversible hepatotoxicity, marked improvement of symptoms was observed after the five-day course of remdesivir was completed. CONCLUSION: Our findings describe the first instance of compassionate use of remdesivir for the treatment of COVID-19 in the setting of end-stage renal disease, acute respiratory distress syndrome, and hypoxemic respiratory failure.

A Case Series of Elizabethkingia meningosepticum Bacteremia in the Cancer Population.

Elizabethkingia meningosepticum (E. meningosepticum) is a ubiquitous microorganism previously known as Chryseobacterium meningosepticum. It is emerging as a pathogen responsible for bacteremia in the immunocompromised population, particularly in cancer patients and those with a history of prolonged hospital stay and frequent instrumentations. A retrospective chart review of all cases over 10 years at the Moffitt Cancer Center showed a total of three patients with E. meningosepticum infection. The first patient (history of multiple myeloma) underwent endoscopy complicated by aspiration pneumonia and positive blood culture for E. meningosepticum infection. The second patient (current acute myelogenous leukemia) had neutropenic fever in the setting of a recent chest port infection. Blood culture from the chest port showed E. meningosepticum. The third patient (history of esophageal adenocarcinoma and acute myelogenous leukemia) had a history of recent pneumonia and cellulitis who presented with recurrent neutropenic fever. Blood culture was positive for E. meningosepticum. E. meningosepticum bacteremia has a high 28-day mortality rate (41%). As these three cases illustrate, early identification of the pathogen along with empiric treatment with a fluoroquinolone and/or minocycline is indicated to reduce morbidity and mortality.

COVID-19 in Immunocompromised Cancer Patients: A Case Series and Review of the Literature.

The global pandemic of the novel coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has presented newfound challenges to the oncology community regarding management of disease progression in immunocompromised and cancer patients. Further, the large influx of COVID-19 patients has overwhelmed healthcare facilities, limited access to intensive care unit beds and ventilators, and canceled elective surgeries causing disruptions to the cancer care continuum and re-organization of oncological care. While it is known that the potential threat of infection is greatest in elderly patients (>60 years of age) and patients with underlying comorbidities, there is still insufficient data to determine the risk of COVID-19 in cancer patients. Given the immunosuppressive status in cancer patients arising from chemotherapy and other comorbidities, management of COVID-19 in this patient population carries a unique set of challenges. We report three cases of COVID-19 in immunocompromised cancer patients and discuss the challenges in preventing, diagnosing, and treating this vulnerable group.

Human Immunodeficiency Virus and Severe Acute Respiratory Syndrome Coronavirus 2 Coinfection: A Systematic Review of the Literature and Challenges.

The concurrence of infection with human immunodeficiency virus (HIV) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of coronavirus disease 2019 (COVID-19), presents an intriguing problem with many uncertainties underlying their pathogenesis. Despite over 96.2 million cases of COVID-19 worldwide as of January 22, 2021, reports of patients coinfected with HIV and SARS-CoV-2 are scarce. It remains unknown whether HIV patients are at a greater risk of infection from SARS-CoV-2, despite their immunocompromised status. We present a systematic review of the literature reporting cases of HIV and SARS-CoV-2 coinfection, and examine trends of clinical outcomes among coinfected patients. We systematically compiled 63 reports of HIV-1 and SARS-CoV-2 coinfection, published as of January 22, 2021. These studies were retrieved through targeted search terms applied to PubMed/Medline and manual search. Despite scattered evidence, reports indicate a favorable prognosis for HIV patients with strict adherence to combined antiretroviral therapy (cART). However, the presence of comorbidities was associated with a poorer prognosis in HIV/SARS-CoV-2 patients, despite cART and viral suppression. Studies were limited by geographic coverage, small sample size, lack of patient details, and short follow-up durations. Although some anti-HIV drugs have shown promising in vitro activity against SARS-CoV-2, there is no conclusive evidence of the clinical efficacy of any anti-HIV drug in the treatment of COVID-19. Further research is needed to explain the under-representation of severe COVID-19 cases among the HIV patient population and to explore the possible protective mechanisms of cART in this vulnerable population.

Disseminated Corynebacterium jeikeium Infection in Cancer Patients.

Corynebacterium jeikeium is a multidrug-resistant gram-positive bacterium of the human skin flora and one of the most clinically important nondiphtherial corynebacteria in the acute care setting. C. jeikeium can cause different forms of infections, especially in immunocompromised patients with underlying risk factors and comorbidities. C. jeikeium was initially described in 1976 as a highly resistant coryneform bacteria causing severe sepsis in patients with hematologic malignancies and profound neutropenia. C. jeikeium infection has also been reported in the setting of endocarditis, septicemia, meningitis, pneumonia, and soft tissue infections. Management of disseminated C. jeikeium infection in immunocompromised cancer patients can be challenging due to its high virulence and rapid skin colonization. We present two cases of disseminated C. jeikeium infection in patients with acute myelogenous leukemia (AML) and underlying comorbidities. Both patients presented with neutropenic fever resistant to initial standard empiric antibiotic therapy.

Complex Regional Pain Syndrome Caused by an Axillary Lipoma.

Complex regional pain syndrome (CRPS), formerly known as reflex sympathetic dystrophy syndrome, is a rare chronic neuro-inflammatory pain condition, which can follow a soft-tissue, bone (type I), or nerve injury (type II) that can be severe and often lasts longer than the original tissue damage. Lipomas impinging on the brachial plexus are rare. To date, there have been no documented cases of CRPS caused by a benign tumor. Here, we report a rare case of CRPS caused by surgical removal of a left axillary lipoma impinging on the brachial plexus. The patient presented with neuropathic pain and hyperalgesia of the left arm, in a non-dermatomal pattern, and pain out of proportion to touch and painful stimulus. Persistent CRPS continued to occur post-operatively for one year without significant change in her pain characteristics. CRPS following elective or emergent surgery to the extremities can pose significant complications to recovery and post-operative care. This condition can be induced through surgery or trauma, which can complicate recovery, impair motor functionality, and cause debilitating pain. Treatment modalities and pathogenesis for CRPS remain obscure and limited, which leads to wide misdiagnosis. Our case highlights the importance of considering CRPS when evaluating differential diagnoses for pre- and post-operative conditions affecting the upper and lower extremities.

Post-operative emergence of acute brachial neuritis following posterior cervical laminectomy with fusion: A case report and review of the literature.

INTRODUCTION: Idiopathic brachial plexus neuritis or neuralgic amyotrophy is a rare neurological condition whose true etiology currently remains unknown. Epidemiologically, the incidence of this condition is exceptionally rare with only 1.6 cases for every 100,000 people (Turner and Parsonage, 1987). Symptoms present an initial acute and sudden pain to the shoulder girdle and upper arm which is followed by a sense of profound weakness and numbness to the upper arm (Parsonage and Turner, 1948). Localized neuropathy within the arm-pit region may also be presented. The pain often exacerbates upon movement of the shoulder. Due to the anatomic location affected and the nature of the clinical symptoms presented, accurate diagnosis of brachial plexus neuritis poses a challenging diagnostic task for physicians due to remarkably similar symptoms expressed by differential diagnoses. PRESENTATION OF CASE: Here, we report the case of a 55-year-old woman who underwent surgery entailing cervical laminectomy with instrumented fusion. She presented with postoperative symptoms of severe pain in the left arm with significant weakness within 24 h after surgery. A diagnosis of brachial plexus neuritis was made based on the symptoms presented and upon review of imaging scans. DISCUSSION: After a six-month follow-up visit, the patient recovered from the brachial neuritis but has residual numbness in the hand. The presentation of this case serves to transmit three fundamental purposes. First, this case serves to establish an intriguing possible association of the post-surgical period of cervical laminectomy with acute brachial neuritis and signifies the importance post-operative linkage with brachial neuritis in general. Second, this case also highlights the importance of close clinical monitoring of patients with unique symptoms within the postoperative follow-up period to ensure successful improvement and accurate diagnosis. CONCLUSION: As an underdiagnosed and relatively obscure condition, this case serves as an imperative reference for physicians to illuminate differential diagnosis of similar symptomatic conditions and also to promote knowledge of brachial plexus neuritis which can lead to an early and precise diagnosis.

Automated Cell Counts on Tissue Sections by Deep Learning and Unbiased Stereology.

In recent decades stereology-based studies have played a significant role in understanding brain aging and developing novel drug discovery strategies for treatment of neurological disease and mental illness. A major obstacle to further progress in a wide range of neuroscience sub-disciplines remains the lack of high-throughput technology for stereology analyses. Though founded on methodologically unbiased principles, commercially available stereology systems still rely on well-trained humans to manually count hundreds of cells within each region of interest (ROI). Even for a simple study with 10 controls and 10 treated animals, cell counts typically require over a month of tedious labor and high costs. Furthermore, these studies are prone to errors and poor reproducibility due to human factors such as subjectivity, variable training, recognition bias, and fatigue. Here we propose a deep neural network-stereology combination to automatically segment and estimate the total number of immunostained neurons on tissue sections. Our three-step approach consists of (1) creating extended-depth-of-field (EDF) images from z-stacks of images (disector stacks); (2) applying an adaptive segmentation algorithm (ASA) to label stained cells in the EDF images (i.e., create masks) for training a convolutional neural network (CNN); and (3) use the trained CNN model to automatically segment and count the total number of cells in test disector stacks using the optical fractionator method. The automated stereology approach shows less than 2% error and over 5× greater efficiency compared to counts by a trained human, without the subjectivity, tedium, and poor precision associated with conventional stereology.