RESUMEN
Deep learning techniques offer improvements in computer-aided diagnosis systems. However, acquiring image domain annotations is challenging due to the knowledge and commitment required of expert pathologists. Pathologists often identify regions in whole slide images with diagnostic relevance rather than examining the entire slide, with a positive correlation between the time spent on these critical image regions and diagnostic accuracy. In this paper, a heatmap is generated to represent pathologists' viewing patterns during diagnosis and used to guide a deep learning architecture during training. The proposed system outperforms traditional approaches based on color and texture image characteristics, integrating pathologists' domain expertise to enhance region of interest detection without needing individual case annotations. Evaluating our best model, a U-Net model with a pre-trained ResNet-18 encoder, on a skin biopsy whole slide image dataset for melanoma diagnosis, shows its potential in detecting regions of interest, surpassing conventional methods with an increase of 20%, 11%, 22%, and 12% in precision, recall, F1-score, and Intersection over Union, respectively. In a clinical evaluation, three dermatopathologists agreed on the model's effectiveness in replicating pathologists' diagnostic viewing behavior and accurately identifying critical regions. Finally, our study demonstrates that incorporating heatmaps as supplementary signals can enhance the performance of computer-aided diagnosis systems. Without the availability of eye tracking data, identifying precise focus areas is challenging, but our approach shows promise in assisting pathologists in improving diagnostic accuracy and efficiency, streamlining annotation processes, and aiding the training of new pathologists.
RESUMEN
With the incidence of skin cancer increasing in the United States, the role of dental providers in noting suspicious skin lesions as well as advising patients in proper sun-protective strategies is more important than ever. However, the public conversation about proper sun protection is often laden with misunderstandings and pervasive myths. This article reviews the most common types of skin cancer, elucidates the differences among the types of UV radiation, and offers updated recommendations for proper sunscreen use along with adjunctive sun-protective strategies.
Asunto(s)
Neoplasias Cutáneas , Protectores Solares , Humanos , Estados Unidos , Protectores Solares/uso terapéutico , Neoplasias Cutáneas/prevención & control , Rayos Ultravioleta/efectos adversosRESUMEN
Although pathologists have their own viewing habits while diagnosing, viewing behaviors leading to the most accurate diagnoses are under-investigated. Digital whole slide imaging has enabled investigators to analyze pathologists' visual interpretation of histopathological features using mouse and viewport tracking techniques. In this study, we provide definitions for basic viewing behavior variables and investigate the association of pathologists' characteristics and viewing behaviors, and how they relate to diagnostic accuracy when interpreting whole slide images. We use recordings of 32 pathologists' actions while interpreting a set of 36 digital whole slide skin biopsy images (5 sets of 36 cases; 180 cases total). These viewport tracking data include the coordinates of a viewport scene on pathologists' screens, the magnification level at which that viewport was viewed, as well as a timestamp. We define a set of variables to quantify pathologists' viewing behaviors such as zooming, panning, and interacting with a consensus reference panel's selected region of interest (ROI). We examine the association of these viewing behaviors with pathologists' demographics, clinical characteristics, and diagnostic accuracy using cross-classified multilevel models. Viewing behaviors differ based on clinical experience of the pathologists. Pathologists with a higher caseload of melanocytic skin biopsy cases and pathologists with board certification and/or fellowship training in dermatopathology have lower average zoom and lower variance of zoom levels. Viewing behaviors associated with higher diagnostic accuracy include higher average and variance of zoom levels, a lower magnification percentage (a measure of consecutive zooming behavior), higher total interpretation time, and higher amount of time spent viewing ROIs. Scanning behavior, which refers to panning with a fixed zoom level, has marginally significant positive association with accuracy. Pathologists' training, clinical experience, and their exposure to a range of cases are associated with their viewing behaviors, which may contribute to their diagnostic accuracy. Research in computational pathology integrating digital imaging and clinical informatics opens up new avenues for leveraging viewing behaviors in medical education and training, potentially improving patient care and the effectiveness of clinical workflow.
RESUMEN
Invasive melanoma, a common type of skin cancer, is considered one of the deadliest. Pathologists routinely evaluate melanocytic lesions to determine the amount of atypia, and if the lesion represents an invasive melanoma, its stage. However, due to the complicated nature of these assessments, inter- and intra-observer variability among pathologists in their interpretation are very common. Machine-learning techniques have shown impressive and robust performance on various tasks including healthcare. In this work, we study the potential of including semantic segmentation of clinically important tissue structure in improving the diagnosis of skin biopsy images. Our experimental results show a 6% improvement in F-score when using whole slide images along with epidermal nests and cancerous dermal nest segmentation masks compared to using whole-slide images alone in training and testing the diagnosis pipeline.
RESUMEN
The number of melanoma diagnoses has increased dramatically over the past three decades, outpacing almost all other cancers. Nearly 1 in 4 skin biopsies is of melanocytic lesions, highlighting the clinical and public health importance of correct diagnosis. Deep learning image analysis methods may improve and complement current diagnostic and prognostic capabilities. The histologic evaluation of melanocytic lesions, including melanoma and its precursors, involves determining whether the melanocytic population involves the epidermis, dermis, or both. Semantic segmentation of clinically important structures in skin biopsies is a crucial step towards an accurate diagnosis. While training a segmentation model requires ground-truth labels, annotation of large images is a labor-intensive task. This issue becomes especially pronounced in a medical image dataset in which expert annotation is the gold standard. In this paper, we propose a two-stage segmentation pipeline using coarse and sparse annotations on a small region of the whole slide image as the training set. Segmentation results on whole slide images show promising performance for the proposed pipeline.
Asunto(s)
Melanoma , Humanos , Melanoma/diagnóstico por imagen , Melanoma/patología , Procesamiento de Imagen Asistido por Computador/métodos , Piel/diagnóstico por imagen , Piel/patología , Epidermis/patología , BiopsiaRESUMEN
OBJECTIVES: Coronavirus disease 2019 (COVID-19) upended in-person medical education, relocating many activities online. We designed a completely virtual pathology rotation to replace our traditional visiting rotation. METHODS: The virtual away rotation was listed in the Visiting Student Application Service (VSAS) and advertised on social media as well as various medical student mailing lists. Nine students were selected to participate in three month-long rotations. The virtual curriculum mirrored our typical in-person clerkship with didactic lectures and daily signout but also included activities exclusive to the virtual rotation such as digitally scanned slide trays and small-group problem-based learning. Anonymous surveys were conducted in which both participants and instructors rated their experiences. RESULTS: Postrotation feedback was overwhelmingly positive from both participants and instructors. Students considered virtual slide sessions as the most effective teaching tool and did not feel hindered overall by lack of in-person experiences. Volunteer trainee instructors indicated the experience improved their teaching and diagnostic skills and expressed interest in teaching future virtual courses. CONCLUSIONS: The success of the virtual away rotation raises consideration for applications beyond the pandemic era and may provide a more level playing field for medical students from underrepresented groups.
Asunto(s)
COVID-19 , Estudiantes de Medicina , Curriculum , Humanos , Pandemias , SARS-CoV-2RESUMEN
BACKGROUND: Histopathologically ambiguous melanocytic lesions lead some pathologists to list multiple diagnostic considerations in the pathology report. The frequency and circumstance of multiple diagnostic considerations remain poorly characterized. METHODS: Two hundred and forty skin biopsy samples were interpreted by 187 pathologists (8976 independent diagnoses) and classified according to a diagnostic/treatment stratification (MPATH-Dx). RESULTS: Multiple diagnoses in different MPATH-Dx classes were used in n = 1320 (14.7%) interpretations, with 97% of pathologists and 91% of cases having at least one such interpretation. Multiple diagnoses were more common for intermediate risk lesions and are associated with greater subjective difficulty and lower confidence. We estimate that 6% of pathology reports for melanocytic lesions in the United States contain two diagnoses of different MPATH-Dx prognostic classes, and 2% of cases are given two diagnoses with significant treatment implications. CONCLUSIONS: Difficult melanocytic diagnoses in skin may necessitate multiple diagnostic considerations; however, as patients increasingly access their health records and retrieve pathology reports (as mandated by US law), uncertainty should be expressed unambiguously.
Asunto(s)
Patólogos , Neoplasias Cutáneas/clasificación , Neoplasias Cutáneas/diagnóstico , Piel/patología , Adulto , Anciano , Biopsia , Femenino , Humanos , Masculino , Melanocitos/patología , Persona de Mediana Edad , Terminología como AsuntoRESUMEN
CONTEXT.: Pathology education must evolve as medical knowledge expands and disruptive technologies emerge. The evolution in pathology teaching practices accelerated as traditional teaching modalities were suspended in March 2020 during the coronavirus disease 2019 (COVID-19) pandemic. OBJECTIVES.: To provide pathologists an overview of established teaching paradigms and practical examples of how these paradigms may be applied to pathology education, emphasizing differences in graduate and undergraduate medical education as well as the challenges and promises of remote learning, as revealed by the COVID-19 pandemic. DATA SOURCES.: Selected peer-reviewed publications representing the field of educational social science. CONCLUSIONS.: Evidence-based methods described in education and social sciences can be effectively deployed in pathology education and especially remote learning, as necessitated by the current COVID-19 pandemic. Understanding established principles, such as cognitive load, competency-based learning, peer-assisted learning, and flipped classrooms may prove useful in developing effective, learner-centric content for pathology education.
Asunto(s)
Educación a Distancia/métodos , Educación Médica/métodos , Patología Clínica/educación , COVID-19 , Curriculum , Educación a Distancia/tendencias , Educación Médica/tendencias , Educación de Postgrado en Medicina , Educación de Pregrado en Medicina , Medicina Basada en la Evidencia , Humanos , Pandemias , Patología Clínica/tendencias , SARS-CoV-2RESUMEN
Diagnosing melanocytic lesions is one of the most challenging areas of pathology with extensive intra- and inter-observer variability. The gold standard for a diagnosis of invasive melanoma is the examination of histopathological whole slide skin biopsy images by an experienced dermatopathologist. Digitized whole slide images offer novel opportunities for computer programs to improve the diagnostic performance of pathologists. In order to automatically classify such images, representations that reflect the content and context of the input images are needed. In this paper, we introduce a novel self-attention-based network to learn representations from digital whole slide images of melanocytic skin lesions at multiple scales. Our model softly weighs representations from multiple scales, allowing it to discriminate between diagnosis-relevant and -irrelevant information automatically. Our experiments show that our method outperforms five other state-of-the-art whole slide image classification methods by a significant margin. Our method also achieves comparable performance to 187 practicing U.S. pathologists who interpreted the same cases in an independent study. To facilitate relevant research, full training and inference code is made publicly available at https://github.com/meredith-wenjunwu/ScATNet.
RESUMEN
CASE PRESENTATION: A 68-year-old man developed an erythematous, papular, pruritic rash on his right thigh 1 month prior to presentation. It subsequently spread to his other extremities and trunk. He also endorsed fevers of > 38.3°C, night sweats, fatigue, shortness of breath, and a dry cough. He was prescribed triamcinolone 0.1% cream for his rash and azithromycin for presumed community-acquired pneumonia, with no improvement in symptoms. He had a history of relapsing polychondritis for which he was prescribed infliximab and low-dose prednisone. He had never smoked tobacco, did not use alcohol or illicit substances, and had no significant travel history.
Asunto(s)
Exantema/etiología , Derrame Pleural/etiología , Policondritis Recurrente/complicaciones , Policondritis Recurrente/diagnóstico , Síndrome de Sweet/complicaciones , Síndrome de Sweet/diagnóstico , Anciano , Exantema/diagnóstico , Exantema/terapia , Humanos , Masculino , Derrame Pleural/diagnóstico , Derrame Pleural/terapia , Policondritis Recurrente/terapia , Síndrome de Sweet/terapiaRESUMEN
Metastatic disease to the oral cavity is rare. Patients with metastasis to the oral cavity may present with swelling, pain, and paresthesia and require evaluation by providers trained in managing pathoses of the oral cavity and surrounding structures. This report describes the case of a 78-year-old man with painful enlargement of the right posterior mandible that caused paresthesia. An open biopsy procedure resulted in significant blood loss and the need for percutaneous needle biopsy. Immunohistochemical analysis was used to make the diagnosis of metastatic clear cell renal cell carcinoma in this patient, whose primary malignancy was previously unknown. Composite resection of the metastatic lesion and reconstruction were performed with the use of virtual surgical planning, an osteomyocutaneous free tissue transfer, and a custom reconstruction plate. This case highlights the importance of dental professionals in the diagnosis and management of lesions of the head and neck and adds to the literature on metastatic lesions to the region.
Asunto(s)
Carcinoma de Células Renales/diagnóstico , Carcinoma de Células Renales/cirugía , Neoplasias Renales , Neoplasias Mandibulares/diagnóstico , Anciano , Biopsia , Humanos , Masculino , Mandíbula/cirugíaRESUMEN
Significant communication occurs between pathologists and clinicians through the dermatopathology report. Our objective was to describe clinician preference about reporting of the margin status of skin biopsies of nonmelanoma skin cancers. An anonymous survey was sent to 243 medical providers who submitted specimens to a single institution university medical center; 50 complete responses from attending-level providers and advance practice providers were received. The majority (96%) of those surveyed indicated margins should be reported on skin biopsies of neoplasms, particularly nonmelanoma skin cancers (basal cell carcinoma 96% and squamous cell carcinoma 92%) and atypical nevi (96%). When asked about particular language used to describe the margin status, some phrasing led to more variance in respondents' clinical management decisions, with 96%-98% of respondents making the same decision when presented with "unambiguous" terms and 58%-84% of respondents making the same decision when presented with "ambiguous" language (P < 0.001). Respondents generally preferred "unambiguous" margin descriptions when shown an involved margin (70% vs. 30%, P < 0.001) but accepted "ambiguous" language when the margin was clearly uninvolved (68% vs. 32%, P = 0.015). Most respondents (88%) desire inclusion of treatment recommendations in dermatopathology reports. Microscopic descriptions were highly utilized, particularly by nondermatology trained clinicians (97% vs. 80%, P = 0.09). Clinicians desire inclusion of margins for skin biopsies in dermatopathology reports, at least in some circumstances. The choice of language used to describe the margin status in dermatopathology reports has important implications for patient care. Margin descriptors that are unclear or ambiguous may lead to more variance in clinical management.
Asunto(s)
Dermatología , Márgenes de Escisión , Patología Quirúrgica , Neoplasias Cutáneas/diagnóstico , Terminología como Asunto , Adulto , Biopsia , Femenino , Humanos , Lenguaje , Masculino , Persona de Mediana EdadRESUMEN
Proliferative index is a prognostic feature of invasive ductal carcinoma of the breast, and has more recently emerged as a predictor of ductal carcinoma in situ (DCIS) local recurrence and progression when used in combination with other predictive markers. Ki67 is the most commonly used immunohistochemical marker of proliferative index. However, high interobserver and interlaboratory variability has been reported, in part due to differences in staining methodologies, positivity thresholds, and approaches to quantification. Phosphohistone-H3 (pHH3) is a marker of mitotic activity that has emerged as a more reliable indicator of proliferation in other neoplasms. Quantification of proliferative index was compared in 48 cases of DCIS using Ki67 and pHH3 immunohistochemistry. A strong linear relationship between Ki67 and pHH3 quantification was observed (P<0.0001, R=0.75). Interobserver concordance was modestly higher for pHH3 than Ki67 proliferative indices. However, positive pHH3 staining was more dichotomous (either negative or uniformly positive) and specific for mitotic activity, and interpretation of pHH3 proliferative indices was significantly faster than that of Ki67. The strong correlation between pHH3 and Ki67 supports the use of this marker as a measure of proliferative activity in DCIS.
Asunto(s)
Biomarcadores de Tumor/metabolismo , Neoplasias de la Mama/metabolismo , Carcinoma Intraductal no Infiltrante/metabolismo , Proliferación Celular , Histonas/metabolismo , Antígeno Ki-67/metabolismo , Fosfoproteínas/metabolismo , Neoplasias de la Mama/patología , Carcinoma Intraductal no Infiltrante/patología , Femenino , Humanos , InmunohistoquímicaRESUMEN
Free-living amebae are ubiquitous in our environment, but rarely cause cutaneous infection. Balamuthia mandrillaris has a predilection for infecting skin of the central face. Infection may be restricted to the skin or associated with life-threatening central nervous system (CNS) involvement. We report a case of a 91-year-old woman, who presented with a non-healing red plaque over her right cheek. Several punch biopsies exhibited non-specific granulomatous inflammation without demonstrable fungi or mycobacteria in histochemical stains. She was treated empirically for granulomatous rosacea, but the lesion continued to progress. A larger incisional biopsy was performed in which amebae were observed in hematoxylin-eosin stained sections. These were retrospectively apparent in the prior punch biopsy specimens. Immunohistochemistry and polymerase chain reaction studies identified the organisms as Balamuthia mandrillaris. Cutaneous infection by B. mandrillaris is a rare condition that is sometimes complicated by life-threatening CNS involvement and which often evades timely diagnosis due to its rarity and nonspecific clinical manifestations. Moreover, these amebae are easily overlooked in histopathologic sections because of their small number and their resemblance to histiocytes. Dermatopathologists should be familiar with the histopathologic appearance of these organisms and include balamuthiasis and other amebic infections in the differential diagnosis of granulomatous dermatitis.