Skin cancer is one of the most serious health issues that humans face. Dermatologists face difficulty in making a skin cancer diagnosis because many skin cancer pigments seem alike. Early detection of skin cancers like Melanoma means a better chance of survival for the patient otherwise it can be life-threatening. For computer vision problems like image classification, deep learning has proven to be the state-of-the-art. There has been a great deal of research into the use of deep learning to automate skin cancer screening. The objective of this paper is to review the state-of-the-art CNN techniques used for Melanoma detection. This paper presents an overview of CNN, followed by analysing the existing work carried out in the area of Melanoma skin cancer detection using Convolution Neural Network (CNN).

The novel coronavirus disease 2019 (COVID-19) has affected more than 1,000,000 people worldwide with the number of cases and deaths surging as of April 4th 2020 per WHO report. Criteria to test is quickly evolving as COVID-19's presentation is better understood; however, lack of adequate testing is still a limitation. While the most common presenting symptoms are fever (44-98%), cough (46-82%), and dyspnea (20-64 %), there are less common symptoms such as diarrhea (10%) (3). Prior criteria for COVID-19 testing excluded those with positive respiratory viral panels; however, a study estimates that about 1 in 5 people infected with COVID-19 are coinfected with another respiratory virus (4). Another study from China showed 5 in 115 patients were co-infected with influenza (2). The following is a case of concomitant COVID-19 and rhino/enterovirus infection presenting predominantly as fever and diarrhea. CASE PRESENTATION: An 85-year-old male with a history of prostate cancer status post prostatectomy, urothelial carcinoma status post resection and chemo-radiation presented with 5 days of ongoing watery, non-bloody diarrhea, decreased appetite, and subjective fever and chills. He reported nausea but denied vomiting, abdominal pain, recent travel, or antibiotic use. He had multiple exposures to family and friends as his wife died the prior week from a COVID-19 negative respiratory illness. He denied cough, congestion, dyspnea, chest pain. Initial labs revealed leukopenia with lymphopenia, thrombocytopenia, elevated CRP and LDH. The patient was also hypoxemic at 85% on room air and was started on supplemental oxygen. Chest x-ray (CXR) showed perihilar and bilateral lower lobe ground glass opacities (GGOs). Viral respiratory panel was positive for rhino/enterovirus and COVID-19 testing was positive. He was treated supportively and discharged within a few days in improved condition. DISCUSSION: This case shows the importance of maintaining a low threshold to test for COVID-19 even in the presence of a positive viral respiratory panel and/or absence of upper respiratory symptoms. Lymphopenia is non-specific however noted in up to 50-60% of patients with COVID-19 (3,6), as can elevations in CRP and LDH. The most common findings on chest X-ray (CXR) are consolidation (47%) and GGOs (33%). CXR abnormalities were more likely peripheral (41%), bilateral (50%), and have lower lobe predominance (50%) (5). CONCLUSIONS: This patient constitutes an atypical presentation of COVID-19 given primary presenting symptom of diarrhea and coinfection with rhino/enterovirus, which challenges prior belief that coinfection should not be tested. With our understanding of COVID-19 rapidly evolving, we can expect to see more changes in screening criteria due to atypical presentations such as this case.

Systemic fibrinolysis has been approved for the treatment of submassive PE since 1990. The use of CDT for acute PE was approved in 2014. In this retrospective quality improvement study at our community based hospital, we aim to examine the difference in tPA and CDT in regards to improvement of right ventricular (RV) dysfunction/cor pulmonale, length of hospital stay, and mortality in patients with submassive PE.METHODS: Data was collected on patients in the hospital who received tPA or CDT for submassive PE from April 2017 through December 2018. The tPA group consisted of four patients who all had extensive clot burden on chest computed tomography angiography (CTA). The CDT group had twelve patients; two of these patients did not have a pre-treatment echocardiogram and were diagnosed with RV strain from CTA so were excluded from the CDT data. First pre and post treatment echocardiograms were reviewed paying attention to the degree of RV dilation as well as the RV systolic pressure (RVSP). In order to standardize the dilation, a grading system was implemented with normal being zero up to massive being four. RESULTS:The tPA group showed an average pre-treatment RV dilation of 3.13 and RVSP of 56.5; post-treatment RV dilation of 0 and RVSP of 36. The CDT group showed an average pre-treatment of RV dilation of 1.54 and RVSP of 52.63; post-treatment RV dilation was 0.14 and RVSP 39.5. When comparing the resultant change between the two groups for RVSP improvement no significance was demonstrated with a two tailed t-test (p ¼ 0.74). Mann-Whitney test was used to look at the difference for RV dilation however this too showed no significance. Mortality overall was noted to be the same between both CDT and tPA groups. Hospital length of stay was similar between CDT and tPA groups at 7 and 7.5 days, respectively and showed no significance (p ¼ 0.867). Both groups there were no major bleeding events.CONCLUSIONS: RV dysfunction is a poor prognostic indicator for early death in patients with acute PE. This study shows that in systemic tPA and CDT groups, there was improvement in RV dilation and RVSP following treatment. When compared to systemic tPA, CDT proved to be similarly effective which is consistent with previous findings. With similar outcomes and hospital stays, the risks of intervention and cost effectiveness of each treatment modality should be considered. Studies show the risk of major bleeding requiring transfusion as high as 10%. An additional finding is not all patients received post-treatment echocardiograms, so assessing for improvement in RV dilation and RVSP was not consistent. CLINICAL IMPLICATIONS:We can apply the PDSA model to initiate a protocol for limited echocardiograms after systemic tPA and CDT treatment of PE to further validate previous findings. Both modalities can be used with no significance between tPA and CDT for the treatment of acute PE.