☆ 3.8 Proceedings Paper

On the Reusability of ISIC Data for Training DL Classifiers Applied on Clinical Skin Images

ARTIFICIAL INTELLIGENCE APPLICATIONS AND INNOVATIONS. AIAI 2022 IFIP WG 12.5 INTERNATIONAL WORKSHOPS (2022)

New Trends in Melanoma Detection Using Neural Networks: A Systematic Review

Dan Popescu et al.

Summary: This paper presents a systematic review of recent advances in using artificial intelligence, especially neural network-based systems, for the early detection and diagnosis of melanoma. The study finds that such systems have shown high accuracy and efficiency in detecting and diagnosing melanoma at the early stages.

SENSORS (2022)

添加到收藏夹

Article Medicine, General & Internal

A Deep Learning Based Framework for Diagnosing Multiple Skin Diseases in a Clinical Environment

Chen-Yu Zhu et al.

Summary: This study aimed to construct a novel framework based on deep learning trained by a dataset that represented the real clinical environment in a tertiary class hospital in China, for better adaptation of the AI application in clinical practice among Asian patients. The results showed that the proposed framework achieved a high level of classification performance with an overall accuracy of 0.948, higher than existing models, and comparable to board-certificated dermatologists in diagnostic tasks.

FRONTIERS IN MEDICINE (2021)

添加到收藏夹

Article Oncology

Original Research A benchmark for neural network robustness in skin cancer classification

Roman C. Maron et al.

Summary: The study aimed to establish a benchmark for skin cancer classification on dermoscopic images to measure classifier robustness to out-of-distribution (OOD) data. By creating three datasets and evaluating the robustness of four different CNN architectures, it was found that all CNNs showed susceptibility to corruptions and perturbations.

EUROPEAN JOURNAL OF CANCER (2021)

添加到收藏夹

Proceedings Paper Computer Science, Artificial Intelligence

Ensembles of Deep Convolutional Neural Networks for Detecting Melanoma in Dermoscopy Images

Melina Tziomaka et al.

Summary: This paper introduces a methodology using the EfficientNet family of convolutional neural networks for the SIIM-ISIC Melanoma Classification Challenge, aiming to detect malignant melanoma from a dataset of 58,457 dermoscopic images of pigmented skin lesions. To address the severe class imbalance, a loss balancing approach is used, and different model input resolutions are considered for images with varying resolution sizes. Lastly, an ensembling strategy of models trained with different activation functions is applied to increase diversity and overall performance.

COMPUTATIONAL COLLECTIVE INTELLIGENCE (ICCCI 2021) (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence