Boundary Distance Loss for Intra-/Extra-meatal Segmentation of Vestibular Schwannoma

Wijethilake, Navodini; Kujawa, Aaron; Dorent, Reuben; Asad, Muhammad; Oviedova, Anna; Vercauteren, Tom; Shapey, Jonathan

doi:10.1007/978-3-031-17899-3_8

Cited by 6 publications

(10 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Following the approach of, 12 we make use of the default 3D full resolution UNet of the nnU-Net framework (3D nnU-Net) 14 in two stages sequentially, to obtain the whole tumour and intra-/extra-meatal segmentation in each stage respectively. Whole tumour masks for training the stage 2 are generated by performing 5-fold cross validation in the stage 1.…”

Section: Deep Learning Based Segmentationmentioning

confidence: 99%

See 1 more Smart Citation

A clinical guideline driven automated linear feature extraction for vestibular schwannoma

Wijethilake,

Connor,

Oviedova

et al. 2024

Medical Imaging 2024: Image Processing

View full text Add to dashboard Cite

Vestibular Schwannoma is a benign brain tumour that grows from one of the balance nerves. Patients may be treated by surgery, radiosurgery or with a conservative "wait-and-scan" strategy. Clinicians typically use manually extracted linear measurements to aid clinical decision making. This work aims to automate and improve this process by using deep learning based segmentation to extract relevant clinical features through computational algorithms. To the best of our knowledge, our study is the first to propose an automated approach to replicate local clinical guidelines. Our deep learning based segmentation provided Dice-scores of 0.8124 ± 0.2343 and 0.8969 ± 0.0521 for extrameatal and whole tumour regions respectively for T2 weighted MRI, whereas 0.8222 ± 0.2108 and 0.9049 ± 0.0646 were obtained for T1 weighted MRI. We propose a novel algorithm to choose and extract the most appropriate maximum linear measurement from the segmented regions based on the size of the extrameatal portion of the tumour. Using this tool, clinicians will be provided with a visual guide and related metrics relating to tumour progression that will function as a clinical decision aid. In this study, we utilize 187 scans obtained from 50 patients referred to a tertiary specialist neurosurgical service in the United Kingdom. The measurements extracted manually by an expert neuroradiologist indicated a significant correlation with the automated measurements (p < 0.0001). Our code is publicly available at https: //github.com/navodini/AutomatedReportGenerationVS.

show abstract

Section: Deep Learning Based Segmentationmentioning

confidence: 99%

“…9 Previous research work has indeed demonstrated that AI tools are technically capable of automatically detecting and segmenting VS 10,11 and even delineating intra-/extra-meatal components. 12 Such tools can serve as a foundation for the automated feature extraction task.…”

Section: Introductionmentioning

confidence: 99%

A clinical guideline driven automated linear feature extraction for vestibular schwannoma

Wijethilake,

Connor,

Oviedova

et al. 2024

Medical Imaging 2024: Image Processing

View full text Add to dashboard Cite

show abstract

“…These models can be applied for the automatic generation of case reports for multidisciplinary team meetings (MDM) (Wijethilake et al, 2023). The reports in their work include multiple automatically generated views of the tumour and the model segmentation and frequently reported tumour measures, such as volume and extrameatal dimensions.…”

Section: Limitations and Future Workmentioning

confidence: 99%

Deep Learning for Automatic Segmentation of Vestibular Schwannoma: A Retrospective Study from Multi-Centre Routine MRI

Kujawa

Dorent

Connor

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

ObjectiveAutomatic segmentation of vestibular schwannoma (VS) from routine clinical MRI can improve clinical workflow, facilitate treatment decisions, and assist patient management. Previously, excellent automatic segmentation results were achieved on datasets of standardised MRI images acquired for stereotactic surgery planning. However, diagnostic clinical datasets are generally more diverse and pose a larger challenge to automatic segmentation algorithms. Here, we show that automatic segmentation of VS on such datasets is also possible with high accuracy.MethodsWe acquired a large multi-centre routine clinical (MC-RC) dataset of 168 patients with a single sporadic VS who were referred from 10 medical sites and consecutively seen at a single centre. Up to three longitudinal MRI exams were selected for each patient. Selection rules based on image modality, resolution orientation, and acquisition timepoint were defined to automatically select contrast-enhanced T1-weighted (ceT1w) images (n=130) and T2-weighted images (n=379). Manual ground truth segmentations were obtained in an iterative process in which segmentations were: 1) produced or amended by a specialized company; and 2) reviewed by one of three trained radiologists; and 3) validated by an expert team. Inter- and intra-observer reliability was assessed on a subset of 10 ceT1w and 41 T2w images. The MC-RC dataset was split randomly into 3 nonoverlapping sets for model training, hyperparameter-tuning and testing in proportions 70/10/20%. We applied deep learning to train our VS segmentation model, based on convolutional neural networks (CNN) within the nnU-Net framework.ResultsOur model achieved excellent Dice scores when evaluated on the MC-RC testing set as well as the public testing set. On the MC-RC testing set, Dice scores were 90.8±4.5% for ceT1w, 86.1±11.6% for T2w and 82.3±18.4% for a combined ceT1w+T2w input.ConclusionsWe developed a model for automatic VS segmentation on diverse multi-centre clinical datasets. The results show that the performance of the framework is comparable to that of human annotators. In contrast, a model trained a publicly available dataset acquired for Gamma Knife stereotactic radiosurgery did not perform well on the MC-RC testing set. The application of our model has the potential to greatly facilitate the management of patients in clinical practice. Our pre-trained segmentation models are made available online. Moreover, we are in the process of making the MC-RC dataset publicly available.

show abstract

“…Early medical imaging research used basic statistical analysis to investigate associations with tumor prognostic factors. Laterally, interest has moved towards using machine learning and deep learning algorithms for tumor segmentation and prognosis analysis ( 17 , 18 ). This motivated us to look at the imaging and analysis techniques used to evaluate extra-axial tumors and how this work has evolved over time to incorporate methodological advancements.…”

Section: Introductionmentioning

confidence: 99%

Imaging biomarkers associated with extra-axial intracranial tumors: a systematic review

et al. 2023

Self Cite

View full text Add to dashboard Cite

Extra-axial brain tumors are extra-cerebral tumors and are usually benign. The choice of treatment for extra-axial tumors is often dependent on the growth of the tumor, and imaging plays a significant role in monitoring growth and clinical decision-making. This motivates the investigation of imaging biomarkers for these tumors that may be incorporated into clinical workflows to inform treatment decisions. The databases from Pubmed, Web of Science, Embase, and Medline were searched from 1 January 2000 to 7 March 2022, to systematically identify relevant publications in this area. All studies that used an imaging tool and found an association with a growth-related factor, including molecular markers, grade, survival, growth/progression, recurrence, and treatment outcomes, were included in this review. We included 42 studies, comprising 22 studies (50%) of patients with meningioma; 17 studies (38.6%) of patients with pituitary tumors; three studies (6.8%) of patients with vestibular schwannomas; and two studies (4.5%) of patients with solitary fibrous tumors. The included studies were explicitly and narratively analyzed according to tumor type and imaging tool. The risk of bias and concerns regarding applicability were assessed using QUADAS-2. Most studies (41/44) used statistics-based analysis methods, and a small number of studies (3/44) used machine learning. Our review highlights an opportunity for future work to focus on machine learning-based deep feature identification as biomarkers, combining various feature classes such as size, shape, and intensity.Systematic Review Registration: PROSPERO, CRD42022306922

show abstract

Boundary Distance Loss for Intra-/Extra-meatal Segmentation of Vestibular Schwannoma

Cited by 6 publications

References 15 publications

A clinical guideline driven automated linear feature extraction for vestibular schwannoma

A clinical guideline driven automated linear feature extraction for vestibular schwannoma

Deep Learning for Automatic Segmentation of Vestibular Schwannoma: A Retrospective Study from Multi-Centre Routine MRI

Imaging biomarkers associated with extra-axial intracranial tumors: a systematic review

Contact Info

Product

Resources

About