Artificial intelligence for dermatopathology: Current trends and the road ahead

Chen, Simon B.; Novoa, Roberto A.

doi:10.1053/j.semdp.2022.01.003

Cited by 21 publications

(21 citation statements)

References 84 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is warranted to confirm our findings with additional studies conducted in other populations or by using a clinical trial design. Some recent studies investigating the application of novel technology of artificial intelligence including machine learning and deep learning to advance our knowledge toward a better understanding and futuristic management of skin cancer and non-cancer skin diseases have shown promising results [54][55][56][57][58][59][60]. It is believed that similar techniques and methods can be applied in the research of clinical risk prediction in the field of dermatology and in helping clinicians identify patients who can be benefited from the use of metformin in the prevention of non-cancer skin diseases in the future.…”

Section: Discussionmentioning

confidence: 99%

Differential Effects of Metformin on Immune-Mediated and Androgen-Mediated Non-Cancer Skin Diseases in Diabetes Patients: A Retrospective Cohort Study

Tseng

2023

Dermatology

View full text Add to dashboard Cite

Background: Metformin’s effects on non-cancer skin diseases are rarely investigated. Objective: To investigate immune-mediated (urticaria, allergic contact dermatitis and psoriasis) and androgen-mediated (acanthosis nigricans, hidradenitis suppurativa and acne) skin diseases associated with metformin use. Methods: Metformin initiators (n=234,585) and non-metformin initiators (n=125,921) within the initial 12 months of antidiabetic drugs prescription during 1999-2009 were followed up until December 31, 2011. Cox regression weighted for propensity score was used to estimate hazard ratios for metformin initiators versus non-metformin initiators in intention-to-treat (ITT) and per-protocol (PP) analyses. Results: For immune-mediated skin diseases, hazard ratios were 0.930 (95% confidence interval: 0.920–0.940) and 0.930 (0.918–0.943) in ITT and PP analyses, respectively; and the hazard ratios for each specific outcome were all significantly below unity. For androgen-mediated skin diseases, the ITT and PP hazard ratios were 1.110 (1.060–1.162) and 0.990 (0.935–1.048), respectively; and all hazard ratios were not significant for each specific outcome except for acne in the ITT analysis (hazard ratio: 1.116, 95% confidence interval: 1.064–1.170). Conclusion: Metformin use is associated with a significantly lower risk of immune-mediated skin diseases but a lack of preventive effect on androgen-mediated skin diseases.

show abstract

Section: Discussionmentioning

confidence: 99%

Differential Effects of Metformin on Immune-Mediated and Androgen-Mediated Non-Cancer Skin Diseases in Diabetes Patients: A Retrospective Cohort Study

Tseng

2023

Dermatology

View full text Add to dashboard Cite

show abstract

“…Deep-learning methods have shown impressive and robust performance on various tasks and hold promise for providing assistant tools in healthcare research including pathology. Dermatopathology research is not an exception in benefiting from the advancement of artificial intelligence [ 30 , 31 , 32 ]. In the time of cancer monitoring and treatment, AI-developed tools have the potential to assist dermatopathologists especially with challenging cases.…”

Section: Discussionmentioning

confidence: 99%

Improving the Diagnosis of Skin Biopsies Using Tissue Segmentation

Nofallah

Mokhtari

et al. 2022

Diagnostics

View full text Add to dashboard Cite

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.

show abstract

“…In recent years, with the development of deep learning in the field of computer vision, the study of medical image processing based on deep learning has become a popular research topic in the field of computer-aided diagnosis [ 12 , 13 , 14 ], and methods using deep learning are gradually being used for the diagnosis and screening of a variety of ophthalmic diseases, however, less research has been conducted on eyelid tumors.…”

Section: Related Workmentioning

confidence: 99%

MAC-ResNet: Knowledge Distillation Based Lightweight Multiscale-Attention-Crop-ResNet for Eyelid Tumors Detection and Classification

Huang

Yao²,

Xing³

et al. 2022

JPM

View full text Add to dashboard Cite

Eyelid tumors are tumors that occur in the eye and its appendages, affecting vision and appearance, causing blindness and disability, and some having a high lethality rate. Pathological images of eyelid tumors are characterized by large pixels, multiple scales, and similar features. Solving the problem of difficult and time-consuming fine-grained classification of pathological images is important to improve the efficiency and quality of pathological diagnosis. The morphology of Basal Cell Carcinoma (BCC), Meibomian Gland Carcinoma (MGC), and Cutaneous Melanoma (CM) in eyelid tumors are very similar, and it is easy to be misdiagnosed among each category. In addition, the diseased area, which is decisive for the diagnosis of the disease, usually occupies only a relatively minor portion of the entire pathology section, and screening the area of interest is a tedious and time-consuming task. In this paper, deep learning techniques to investigate the pathological images of eyelid tumors. Inspired by the knowledge distillation process, we propose the Multiscale-Attention-Crop-ResNet (MAC-ResNet) network model to achieve the automatic classification of three malignant tumors and the automatic localization of whole slide imaging (WSI) lesion regions using U-Net. The final accuracy rates of the three classification problems of eyelid tumors on MAC-ResNet were 96.8%, 94.6%, and 90.8%, respectively.

show abstract

Artificial intelligence for dermatopathology: Current trends and the road ahead

Cited by 21 publications

References 84 publications

Differential Effects of Metformin on Immune-Mediated and Androgen-Mediated Non-Cancer Skin Diseases in Diabetes Patients: A Retrospective Cohort Study

Differential Effects of Metformin on Immune-Mediated and Androgen-Mediated Non-Cancer Skin Diseases in Diabetes Patients: A Retrospective Cohort Study

Improving the Diagnosis of Skin Biopsies Using Tissue Segmentation

MAC-ResNet: Knowledge Distillation Based Lightweight Multiscale-Attention-Crop-ResNet for Eyelid Tumors Detection and Classification

Contact Info

Product

Resources

About