Exploring convolutional neural networks with transfer learning for diagnosing Lyme disease from skin lesion images

  title={Exploring convolutional neural networks with transfer learning for diagnosing Lyme disease from skin lesion images},
  author={Sk. Imran Hossain and Jocelyn de Go{\"e}r de Herve and Md Shahriar Hassan and Delphine Martineau and Evelina Petrosyan and Violaine Corbin and Jean Beytout and Isabelle Lebert and Jonas Durand and Irene Carravieri and Annick Brun-Jacob and Pascale Frey-Klett and Elisabeth Baux and C{\'e}line Cazorla and Carole Eldin and Yves Hansmann and Sol{\`e}ne Patrat-Delon and Thierry Prazuck and Alice Raffetin and Pierre Tattevin and Gwena{\"e}l Vourc'h and Olivier Lesens and Engelbert Mephu Nguifo},
  journal={Computer methods and programs in biomedicine},
4 Citations

Expert Opinion Elicitation for Assisting Deep Learning based Lyme Disease Classifier with Patient Data

The elicited probability scores can be utilized to make image based deep learning Lyme disease pre-scanners robust and assist the deep learning model with a probability score calculated from patient data.

Early Diagnosis of Lyme Disease by Recognizing Erythema Migrans Skin Lesion from Images Utilizing Deep Learning Techniques

This work extensively studied the effectiveness of convolutional neural networks for identifying Lyme dis-ease from images to deal with lack of data, multimodal learning incorporating expert opinion elicitation, and automation of skin hair mask generation.

Classification of Skin Lesion through Active Learning Strategies.



Skin Image Analysis for Erythema Migrans Detection and Automated Lyme Disease Referral

A pre-screener using a Deep Convolutional Neural Network that classifies EM vs. other conditions, including either control/unaffected skin, or skin presenting with other confuser lesions including those for skin cancer.

Dermatologist-level classification of skin cancer with deep neural networks

This work demonstrates an artificial intelligence capable of classifying skin cancer with a level of competence comparable to dermatologists, trained end-to-end from images directly, using only pixels and disease labels as inputs.

Deep neural networks show an equivalent and often superior performance to dermatologists in onychomycosis diagnosis: Automatic construction of onychomycosis datasets by region-based convolutional deep neural network

By training with a dataset comprising 49,567 images, this study achieved a diagnostic accuracy for onychomycosis using deep learning that was superior to that of most of the dermatologists who participated in this study.

The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions

The HAM10000 (“Human Against Machine with 10000 training images”) dataset is released, which consists of 10015 dermatoscopic images which are released as a training set for academic machine learning purposes and are publicly available through the ISIC archive.

Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists

For the first time, a CNN's diagnostic performance with a large international group of 58 dermatologists, including 30 experts, was compared and most dermatologists were outperformed by the CNN.

Data Augmentation for Skin Lesion Analysis

The results confirm the importance of data augmentation in both training and testing and show that it can lead to more performance gains than obtaining new images.

Supervised Visual System for Recognition of Erythema Migrans, an Early Skin Manifestation of Lyme Borreliosis

A novel visual system for recognition of erythema migrans is presented based on new technology of smartphones and it is found that the results obtained with GrowCut method are better than those obtained with Random Walker method.

A deep learning system for differential diagnosis of skin diseases

A deep learning system able to identify the most common skin conditions may help clinicians in making more accurate diagnoses in routine clinical practice.