Longitudinal Assessment of Posttreatment Diffuse Glioma Tissue Volumes with Three-dimensional Convolutional Neural Networks

Rudie, Jeffrey D.; Calabrese, Evan; Saluja, Rachit; Weiß, David; Colby, John B.; Hess, Christopher P.; Rauschecker, Andreas M.; Sugrue, Leo P.; Villanueva-Meyer, Javier

doi:10.1148/ryai.210243

Cited by 10 publications

(11 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Discussionmentioning

confidence: 99%

“…Few automatic segmentation papers, all of which aimed to measure longitudinal change in tumor burden to assess treatment response, have reported the performance of models in the post-treatment setting [12][13][14]. Rudie et al [12] focused on proposing a solution for assessing change in tumor size (progressed vs not) by training a model on subtracted images in two consecutive timepoints to detect longitudinal change in a cohort of patients with diffuse gliomas. Although the context of their work was not directly com parable to ours, as part of their analysis the authors trained a baseline model on post-treatment images and reached mean Dice coefficients of 0.85 in Edema and 0.71 in active tumor regions.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Towards Longitudinal Glioma Segmentation: Evaluating combined pre- and post-treatment MRI training data for automated tumor segmentation using nnU-Net

Ranjbar

Singleton

Curtin

et al. 2023

Preprint

View full text Add to dashboard Cite

Identification of key phenotypic regions such as necrosis, contrast enhancement, and edema on magnetic resonance imaging (MRI) is important for understanding disease evolution and treatment response in patients with glioma. Manual delineation is time intensive and not feasible for a clinical workflow. Automating phenotypic region segmentation overcomes many issues with manual segmentation, however, current glioma segmentation datasets focus on pre-treatment, diagnostic scans, where treat-ment effects and surgical cavities are not present. Thus, existing automatic segmenta-tion models are not applicable to post-treatment imaging that is used for longitudinal evaluation of care. Here, we present a comparison of three-dimensional convolutional neural networks (nnU-Net architecture) trained on large temporally defined pre-treatment, post-treatment, and mixed cohorts. We used a total of 1563 imaging timepoints from 854 patients curated from 13 different institutions as well as diverse public data sets to understand the capabilities and limitations of automatic segmenta-tion on glioma images with different phenotypic and treatment appearance. We as-sessed the performance of models using Dice coefficients on test cases from each group comparing predictions with manual segmentations generated by trained techni-cians. We demonstrate that training a combined model can be as effective as models trained on just one temporal group. The results highlight the importance of a diverse training set, that includes images from the course of disease and with effects from treatment, in the creation of a model that can accurately segment glioma MRIs at multiple treatment time points.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Towards Longitudinal Glioma Segmentation: Evaluating combined pre- and post-treatment MRI training data for automated tumor segmentation using nnU-Net

Ranjbar

Singleton

Curtin

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Pathologies ranged from tumors to leukoencephalopathy to chronic small vessel ischemia. The default FLAIR U-Net to detect new lesions was trained on 198 patients with brain tumors with consecutive imaging and manual segmentations delineating areas of change on FLAIR imaging ( Rudie et al, 2022 ). The default enhancement U-Net was trained on 463 MR studies demonstrating abnormally enhancing metastatic tumors (“metastases”) from the University of California, San Francisco ( Rudie et al, 2021 ).…”

Section: Methodsmentioning

confidence: 99%

“…Training was performed for 30 epochs with a standard cross-entropy loss, Adam optimizer, and learning rate of 10 −5 ; further details on architecture and training process are described in Duong et al (2019) . To develop MS-specific models, we used MS training data to fine-tune our default disease-invariant FLAIR model ( Duong et al, 2019 ), glioma-specific new FLAIR signal model ( Rudie et al, 2022 ), and metastases-specific enhancement model ( Rudie et al, 2021 ). We compared these fine-tuned models with de-novo models trained with the same data.…”

Section: Methodsmentioning

confidence: 99%

3D U-Net for automated detection of multiple sclerosis lesions: utility of transfer learning from other pathologies

Wahlig,

Nedelec,

Weiss

et al. 2023

Front. Neurosci.

Self Cite

View full text Add to dashboard Cite

Background and purposeDeep learning algorithms for segmentation of multiple sclerosis (MS) plaques generally require training on large datasets. This manuscript evaluates the effect of transfer learning from segmentation of another pathology to facilitate use of smaller MS-specific training datasets. That is, a model trained for detection of one type of pathology was re-trained to identify MS lesions and active demyelination.Materials and methodsIn this retrospective study using MRI exams from 149 patients spanning 4/18/2014 to 7/8/2021, 3D convolutional neural networks were trained with a variable number of manually-segmented MS studies. Models were trained for FLAIR lesion segmentation at a single timepoint, new FLAIR lesion segmentation comparing two timepoints, and enhancing (actively demyelinating) lesion segmentation on T1 post-contrast imaging. Models were trained either de-novo or fine-tuned with transfer learning applied to a pre-existing model initially trained on non-MS data. Performance was evaluated with lesionwise sensitivity and positive predictive value (PPV).ResultsFor single timepoint FLAIR lesion segmentation with 10 training studies, a fine-tuned model demonstrated improved performance [lesionwise sensitivity 0.55 ± 0.02 (mean ± standard error), PPV 0.66 ± 0.02] compared to a de-novo model (sensitivity 0.49 ± 0.02, p = 0.001; PPV 0.32 ± 0.02, p < 0.001). For new lesion segmentation with 30 training studies and their prior comparisons, a fine-tuned model demonstrated similar sensitivity (0.49 ± 0.05) and significantly improved PPV (0.60 ± 0.05) compared to a de-novo model (sensitivity 0.51 ± 0.04, p = 0.437; PPV 0.43 ± 0.04, p = 0.002). For enhancement segmentation with 20 training studies, a fine-tuned model demonstrated significantly improved overall performance (sensitivity 0.74 ± 0.06, PPV 0.69 ± 0.05) compared to a de-novo model (sensitivity 0.44 ± 0.09, p = 0.001; PPV 0.37 ± 0.05, p = 0.001).ConclusionBy fine-tuning models trained for other disease pathologies with MS-specific data, competitive models identifying existing MS plaques, new MS plaques, and active demyelination can be built with substantially smaller datasets than would otherwise be required to train new models.

show abstract

“…While most published AI algorithms for brain tumor segmentation focus on preoperative MRI, a number of studies have evaluated the application of AI models in posttreatment settings and achieved comparable performance as compared to human experts [27,82 ▪▪ ,83]. Segmentation of posttreatment imaging findings allows for more precise quantification of small differences between timepoints, improving sensitivity for detection of small interval changes that may otherwise be missed.…”

Section: Posttreatment Segmentation and Surveillancementioning

confidence: 99%

Artificial intelligence in neuroimaging of brain tumors: reality or still promise?

Pan,

Huang

2023

Current Opinion in Neurology

View full text Add to dashboard Cite

Purpose of review To provide an updated overview of artificial intelligence (AI) applications in neuro-oncologic imaging and discuss current barriers to wider clinical adoption. Recent findings A wide variety of AI applications in neuro-oncologic imaging have been developed and researched, spanning tasks from pretreatment brain tumor classification and segmentation, preoperative planning, radiogenomics, prognostication and survival prediction, posttreatment surveillance, and differentiating between pseudoprogression and true disease progression. While earlier studies were largely based on data from a single institution, more recent studies have demonstrated that the performance of these algorithms are also effective on external data from other institutions. Nevertheless, most of these algorithms have yet to see widespread clinical adoption, given the lack of prospective studies demonstrating their efficacy and the logistical difficulties involved in clinical implementation. Summary While there has been significant progress in AI and neuro-oncologic imaging, clinical utility remains to be demonstrated. The next wave of progress in this area will be driven by prospective studies measuring outcomes relevant to clinical practice and go beyond retrospective studies which primarily aim to demonstrate high performance.

show abstract

Longitudinal Assessment of Posttreatment Diffuse Glioma Tissue Volumes with Three-dimensional Convolutional Neural Networks

Cited by 10 publications

References 14 publications

Towards Longitudinal Glioma Segmentation: Evaluating combined pre- and post-treatment MRI training data for automated tumor segmentation using nnU-Net

Towards Longitudinal Glioma Segmentation: Evaluating combined pre- and post-treatment MRI training data for automated tumor segmentation using nnU-Net

3D U-Net for automated detection of multiple sclerosis lesions: utility of transfer learning from other pathologies

Artificial intelligence in neuroimaging of brain tumors: reality or still promise?

Contact Info

Product

Resources

About