Dataset Information

Models Genesis: Generic Autodidactic Models for 3D Medical Image Analysis.

ABSTRACT: Transfer learning from natural image to medical image has established as one of the most practical paradigms in deep learning for medical image analysis. However, to fit this paradigm, 3D imaging tasks in the most prominent imaging modalities (e.g., CT and MRI) have to be reformulated and solved in 2D, losing rich 3D anatomical information and inevitably compromising the performance. To overcome this limitation, we have built a set of models, called Generic Autodidactic Models, nicknamed Models Genesis, because they are created ex nihilo (with no manual labeling), self-taught (learned by self-supervision), and generic (served as source models for generating application-specific target models). Our extensive experiments demonstrate that our Models Genesis significantly outperform learning from scratch in all five target 3D applications covering both segmentation and classification. More importantly, learning a model from scratch simply in 3D may not necessarily yield performance better than transfer learning from ImageNet in 2D, but our Models Genesis consistently top any 2D approaches including fine-tuning the models pre-trained from ImageNet as well as fine-tuning the 2D versions of our Models Genesis, confirming the importance of 3D anatomical information and significance of our Models Genesis for 3D medical imaging. This performance is attributed to our unified self-supervised learning framework, built on a simple yet powerful observation: the sophisticated yet recurrent anatomy in medical images can serve as strong supervision signals for deep models to learn common anatomical representation automatically via self-supervision. As open science, all pre-trained Models Genesis are available at https://github.com/MrGiovanni/ModelsGenesis.

SUBMITTER: Zhou Z

PROVIDER: S-EPMC7405596 | biostudies-literature | 2019 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Models Genesis: Generic Autodidactic Models for 3D Medical Image Analysis.

Zhou Zongwei Z Sodha Vatsal V Siddiquee Md Mahfuzur Rahman MMR Feng Ruibin R Tajbakhsh Nima N Gotway Michael B MB Liang Jianming J

Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention 20191010

Transfer learning from natural image to medical image has established as one of the most practical paradigms in deep learning for medical image analysis. However, to fit this paradigm, 3D imaging tasks in the most prominent imaging modalities (e.g., CT and MRI) have to be reformulated and solved in 2D, losing rich 3D anatomical information and inevitably compromising the performance. To overcome this limitation, we have built a set of models, called Generic Autodidactic Mode ...[more]

PMID: 32766570

Similar Datasets

Project description:Building automated cancer screening systems based on image analysis is currently a hot topic in computer vision and medical imaging community. One of the biggest challenges of such systems, especially those using state-of-the-art deep learning techniques, is that they usually require a large amount of training data to be accurate. However, in the medical field, the confidentiality of the data and the need for medical expertise to label them significantly reduce the amount of training data available. A common practice to overcome this problem is to apply data set augmentation techniques to artificially increase the size of the training data set. Classical data set augmentation methods such as geometrical or color transformations are efficient but still produce a limited amount of new data. Hence, there has been interest in data set augmentation methods using generative models able to synthesize a wider variety of new data. VitaDX is actually developing an automated bladder cancer screening system based on the analysis of cell images contained in urinary cytology digital slides. Currently, the number of available labeled cell images is limited and therefore exploitation of the full potential of deep learning techniques is not possible. In an attempt to increase the number of labeled cell images, a new generic generator for 2D cell images has been developed and is described in this article. This framework combines previous works on cell image generation and a recent style transfer method referred to as doodle-style transfer in this article. To the best of our knowledge, we are the first to use a doodle-style transfer method for synthetic cell image generation. This framework is quite modular and could be applied to other cell image generation problems. A statistical evaluation has shown that features of real and synthetic cell images followed roughly the same distribution. Finally, the realism of the synthetic cell images has been assessed through a visual evaluation performed with the help of medical experts. © 2019 The Authors. Cytometry Part A published by Wiley Periodicals, Inc. on behalf of International Society for Advancement of Cytometry.

Project description:Background/objectiveArtificial bone models (ABMs) are used in orthopaedics for research of biomechanics, development of implants and educational purposes. Most of the commercially available ABMs approximate the morphology of Europeans, but they may not depict the Asian anatomy. Therefore, our aim was to develop the first Asian ABM of the pelvis and compare it with the existing pelvic ABM (Synbone®; Caucasian male).MethodsOne hundred clinical computed tomography (CTs) of adult pelvises (male n = 50, female n = 50) of Malay, Chinese and Indian descent were acquired. CTs were segmented and defined landmarks were placed. Three 3D statistical pelvic model and mean models (overall, male, female) were generated. Anatomical variations were analysed using principal component analysis. To measure gender-related differences and differences to the existing ABM, distances between the anterior superior iliac spines (ASIS), the anterior inferior iliac spines (AIIS), the promontory and the symphysis (conjugate vera, CV) as well as the ischial spines (diameter transversa, DT) were quantified.ResultsPrincipal component analysis displayed large variability regarding the pelvic shape and size. Female and male statistical models were similar in ASIS (225 ± 20; 227 ± 13 mm; P = 0.4153) and AIIS (185 ± 11; 187 ± 10 mm; P = 0.3982) and differed in CV (116 ± 10; 105 ± 10 mm; P < 0.0001) and DT (105 ± 7; 88 ± 8 mm; P < 0.0001). Comparing the unisex mean model with the pre-existing ABM, the ASIS (226; 275 mm; P < 0.0001), the AIIS (186; 209 mm; P < 0.0001) and the CV (111; 105 mm; P < 0.0001) differed significantly. Both models were similar regarding DT (97; 95 mm; P = 0.6927). The analysis revealed notable gender- and size-dependent anatomical variations within the Asian population. Chinese, Malay and Indian descents did not differ notably. The overall Asian model was smaller than the existing ABM.The translation potential of this articleOwing to the large differences between the Asian ABM and the pre-existing ABM, as well as differences between genders, the use of an Asian- and gender-specific ABM is important to consider in research, biomechanics and implant development for this population.

Dataset Information

Models Genesis: Generic Autodidactic Models for 3D Medical Image Analysis.

Publications

Models Genesis: Generic Autodidactic Models for 3D Medical Image Analysis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets