Comparison of Three Deep Learning Models in Accurate Classification of 770 Dermoscopy Skin Lesion Images.

Accurately determining and classifying different types of skin cancers is critical for early diagnosis. In this work, we propose a novel use of deep learning for classification of benign and malignant skin lesions using dermoscopy images. We obtained 770 de-identified dermoscopy images from the University of Missouri (MU) Healthcare. We created three unique image datasets that contained the original images and images obtained after applying a hair removal algorithm. We trained three popular deep learning models, namely, ResNet50, DenseNet121, and Inception-V3. We evaluated the accuracy and the area under the curve (AUC) receiver operating characteristic (ROC) for each model and dataset. DenseNet121 achieved the best accuracy (80.52%) and AUC ROC score (0.81) on the third dataset. For this dataset, the sensitivity and specificity were 0.80 and 0.81, respectively. We also present the SHAP (SHapley Additive exPlanations) values for the predictions made by different models to understand their interpretability.

It's Not Always Easy: Cancer Survivorship Care in Primary Care Settings.

By 2040, an anticipated 26.1 million people with a history of cancer will be part of the healthcare system. The purpose of this study was to explore Missouri-based non-oncology clinicians' perspectives on caring for patients with a history of cancer to identify needs of rural-based clinicians to optimize their patients' survivorship care. Using an interpretive qualitative descriptive approach, we conducted semi-structured interviews with 17 non-oncology clinicians. We encouraged clinicians to discuss their approach to caring for patients with a history of cancer and invited them to talk about what might help them increase their knowledge of survivorship care best practices. Through interpretive qualitative descriptive analysis methods including first level coding and constant comparison, we found there is consensus that cancer survivorship care is important; however, training that now guides our clinicians occurred mostly during residency, if at all. Clinicians relied on previous patient encounters and oncology notes combined with their patients' personal account of treatment history to inform the best next steps. Clinicians expressed strong interest in having a simple protocol of their patient's treatment with prompts of known long-term cancer treatment-related effects and a patient-centric follow-up monitoring schedule (mandatory vs recommended vs optional). Clinicians expressed interest in educational opportunities about cancer care and ability for curbside consults with oncologists. They consistently noted the limited resources available in rural areas and that rural patients may have different preferences and approaches to cancer survivorship. There is a clear opportunity to improve non-oncology clinicians' knowledge of the needs of people with a history of cancer as well as their own knowledge base and self-efficacy, especially in rural settings.

Melanoma Extension for Community Healthcare Outcomes: A Feasibility Study of Melanoma Screening Implementation in Primary Care Settings.

Introduction Melanoma incidence rates are rising faster than the rates of any other malignancy. As a major global public health concern, melanoma can be identified by a visual exam not requiring expensive invasive procedures. However, non-dermatologists lack specialized training and skills to identify high-risk patients and implement melanoma skin screenings during regular exams. Most patients from rural and underserved areas have inadequate access to specialty dermatologic care, which can potentially lead to later-stage melanomas and poor patient outcomes. The objective of this study was to identify facilitators and barriers to the implementation of risk surveys and melanoma skin screenings in primary care settings through live interactive education and the telementoring project - Melanoma ECHO (Extension for Community Healthcare Outcomes).  Methods This cross-sectional study was designed with theoretical concepts from dissemination and implementation research. Monthly Melanoma ECHO sessions were integrated into an ongoing Dermatology ECHO at the University of Missouri, Columbia, Missouri, USA, from April 2018 to February 2019. Ten primary care providers, medical doctors/doctors of osteopathic medicine (MDs/DOs), nurse practitioners (NPs), and physician assistants (PAs), from across Missouri participated. Eleven virtual monthly melanoma-related didactics and case-based discussions were provided to participants. Information regarding risk factors, risk surveys, and screening techniques was provided. Ongoing telementoring and guidance were also provided for de-identified real-life patient cases. The main outcomes and measures of the study were to identify the facilitators and barriers of risk survey and melanoma skin screenings in primary care settings and to quantify the number of high-risk patients identified by participating providers and the number of new melanomas detected by visual exams during the study period. Results The primary reason why six out of 10 providers reported participation in Melanoma ECHO was that implementing melanoma skin screenings in their practice was made easier as it increased their confidence. Nine providers reported increased knowledge, and eight cited professional networking as other facilitators. The main perceived barrier to melanoma skin screening was lack of administrative and nursing support, and six providers indicated that lack of time to incorporate skin exams was also a barrier. Combined, ten participants reported identifying 976 high-risk patients during the study period and detecting 36 new melanomas. Discussion and conclusion Our findings indicate that primary care providers may benefit from attending regularly scheduled and focused specialized telementoring sessions, such as Melanoma ECHO. Ongoing support from specialists may help providers practicing in rural and isolated areas with the successful integration of risk surveys and melanoma skin screenings in primary care settings. Further Melanoma ECHO sessions with a more diverse group of primary care providers are needed to better understand the generalizability of the results.

Accelerating Care Through ECHO: Case Examples from the Field.

In this article, we describe three life-changing patient cases demonstrating high-quality and timely care they received in their communities, thanks to the Show-Me ECHO project. Early autism diagnosis, a potentially deadly tumor manifesting as a benign-looking rash, a recalcitrant case of hepatitis C: rural and underserved Missourians now have access to state-of-the-art care through their local providers receiving interdisciplinary telementoring on evidence based practices.

Evaluation of Patient Perceptions With Orthopedic Oncology Telehealth: A Pilot Project.

Telehealth has been used for decades to improve access to care for rural and underserved patients. The adoption of telehealth in orthopedic oncology is novel and expected to positively impact patient access and compliance. However, no previous evaluation has been published of this expected impact. The objective of this pilot project was to evaluate patients' perceptions regarding orthopedic oncology telehealth services. A 13-question satisfaction survey was distributed to patients who used tele-orthopedic oncology. Fifteen respondents (a response rate of 42%) reported satisfaction with services at 9.7 of 10. Median travel distance to the nearest in-person orthopedic oncologist was greater than 150 miles (241 km). These results are consistent with the previous findings of high satisfaction with telehealth in other specialties. Health care organizations are likely to benefit from offering telehealth to orthopedic oncology patients with limited access.

Skin cancer screening: recommendations for data-driven screening guidelines and a review of the US Preventive Services Task Force controversy.

Melanoma is usually apparent on the skin and readily detected by trained medical providers using a routine total body skin examination, yet this malignancy is responsible for the majority of skin cancer-related deaths. Currently, there is no national consensus on skin cancer screening in the USA, but dermatologists and primary care providers are routinely confronted with making the decision about when to recommend total body skin examinations and at what interval. The objectives of this paper are: to propose rational, risk-based, data-driven guidelines commensurate with the US Preventive Services Task Force screening guidelines for other disorders; to compare our proposed guidelines to recommendations made by other national and international organizations; and to review the US Preventive Services Task Force's 2016 Draft Recommendation Statement on skin cancer screening.