Significant advances have been made towards building accurate automatic segmentation systems for a variety of biomedical applications using machine learning. However, the performance of these systems ...often degrades when they are applied on new data that differ from the training data, for example, due to variations in imaging protocols. Manually annotating new data for each test domain is not a feasible solution. In this work we investigate unsupervised domain adaptation using adversarial neural networks to train a segmentation method which is more invariant to differences in the input data, and which does not require any annotations on the test domain. Specifically, we learn domain-invariant features by learning to counter an adversarial network, which attempts to classify the domain of the input data by observing the activations of the segmentation network. Furthermore, we propose a multi-connected domain discriminator for improved adversarial training. Our system is evaluated using two MR databases of subjects with traumatic brain injuries, acquired using different scanners and imaging protocols. Using our unsupervised approach, we obtain segmentation accuracies which are close to the upper bound of supervised domain adaptation.
End-stage renal failure is most commonly caused by the obesity-related diseases, diabetes mellitus and essential hypertension, and is best treated with renal transplantation. Obesity may contribute ...to poor patient and graft survival, and is an exclusion criterion in some renal transplant programs. Diet and exercise programs have not proven to be effective for weight loss before transplantation, and bariatric surgery in any form has not been used in this setting before. We report three morbidly obese patients who underwent laparoscopic adjustable gastric banding to meet the criteria for renal transplantation and subsequently were successfully transplanted.
Background: Degenerative cervical myelopathy (DCM) is the most common cause of adult spinal cord dysfunction globally. Associated neurological symptoms and signs have historically been explained by ...pathobiology within the cervical spine. However, recent advances in imaging have shed light on numerous brain changes in patients with DCM, and it is hypothesised that these changes contribute to DCM pathogenesis. The aetiology, significance, and distribution of these supraspinal changes is currently unknown. The objective was therefore to synthesise all current evidence on brain changes in DCM. Methods: A systematic review was performed. Cross-sectional and longitudinal studies with magnetic resonance imaging on a cohort of patients with DCM were eligible. PRISMA guidelines were followed. MEDLINE and Embase were searched to 28th August 2023. Duplicate title/abstract screening, data extraction and risk of bias assessments were conducted. A qualitative synthesis of the literature is presented as per the Synthesis Without Meta-Analysis (SWiM) reporting guideline. The review was registered with PROSPERO (ID: CRD42022298538). Findings: Of the 2014 studies that were screened, 47 studies were identified that used MRI to investigate brain changes in DCM. In total, 1500 patients with DCM were included in the synthesis, with a mean age of 53 years. Brain alterations on MRI were associated with DCM both before and after surgery, particularly within the sensorimotor network, visual network, default mode network, thalamus and cerebellum. Associations were commonly reported between brain MRI alterations and clinical measures, particularly the Japanese orthopaedic association (JOA) score. Risk of bias of included studies was low to moderate. Interpretation: The rapidly expanding literature provides mounting evidence for brain changes in DCM. We have identified key structures and pathways that are altered, although there remains uncertainty regarding the directionality and clinical significance of these changes. Future studies with greater sample sizes, more detailed phenotyping and longer follow-up are now needed. Funding: ODM is supported by an Academic Clinical Fellowship at the University of Cambridge. BMD is supported by an NIHR Clinical Doctoral Fellowship at the University of Cambridge (NIHR300696). VFJN is supported by an NIHR Rosetrees Trust Advanced Fellowship (NIHR302544). This project was supported by an award from the Rosetrees Foundation with the Storygate Trust (A2844).
Segmentation models for brain lesions in MRI are commonly developed for a specific disease and trained on data with a predefined set of MRI modalities. Each such model cannot segment the disease ...using data with a different set of MRI modalities, nor can it segment any other type of disease. Moreover, this training paradigm does not allow a model to benefit from learning from heterogeneous databases that may contain scans and segmentation labels for different types of brain pathologies and diverse sets of MRI modalities. Additionally, the sensitivity of patient data often prevents centrally aggregating data, necessitating a decentralized approach. Is it feasible to use Federated Learning (FL) to train a single model on client databases that contain scans and labels of different brain pathologies and diverse sets of MRI modalities? We demonstrate promising results by combining appropriate, simple, and practical modifications to the model and training strategy: Designing a model with input channels that cover the whole set of modalities available across clients, training with random modality drop, and exploring the effects of feature normalization methods. Evaluation on 7 brain MRI databases with 5 different diseases shows that such FL framework can train a single model that is shown to be very promising in segmenting all disease types seen during training. Importantly, it is able to segment these diseases in new databases that contain sets of modalities different from those in training clients. These results demonstrate, for the first time, the feasibility and effectiveness of using Federated Learning to train a single 3D segmentation model on decentralised data with diverse brain diseases and MRI modalities, a necessary step towards leveraging heterogeneous real-world databases. Code will be made available at: https://github.com/FelixWag/FL-MultiDisease-MRI
Proceedings of Machine Learning Research, MIDL 2024 Models for segmentation of brain lesions in multi-modal MRI are commonly
trained for a specific pathology using a single database with a predefined ...set
of MRI modalities, determined by a protocol for the specific disease. This work
explores the following open questions: Is it feasible to train a model using
multiple databases that contain varying sets of MRI modalities and annotations
for different brain pathologies? Will this joint learning benefit performance
on the sets of modalities and pathologies available during training? Will it
enable analysis of new databases with different sets of modalities and
pathologies? We develop and compare different methods and show that promising
results can be achieved with appropriate, simple and practical alterations to
the model and training framework. We experiment with 7 databases containing 5
types of brain pathologies and different sets of MRI modalities. Results
demonstrate, for the first time, that joint training on multi-modal MRI
databases with different brain pathologies and sets of modalities is feasible
and offers practical benefits. It enables a single model to segment pathologies
encountered during training in diverse sets of modalities, while facilitating
segmentation of new types of pathologies such as via follow-up fine-tuning. The
insights this study provides into the potential and limitations of this
paradigm should prove useful for guiding future advances in the direction. Code
and pretrained models: https://github.com/WenTXuL/MultiUnet
Cranial implant design is a challenging task, whose accuracy is crucial in the context of cranioplasty procedures. This task is usually performed manually by experts using computer-assisted design ...software. In this work, we propose and evaluate alternative automatic deep learning models for cranial implant reconstruction from CT images. The models are trained and evaluated using the database released by the AutoImplant challenge, and compared to a baseline implemented by the organizers. We employ a simulated virtual craniectomy to train our models using complete skulls, and compare two different approaches trained with this procedure. The first one is a direct estimation method based on the UNet architecture. The second method incorporates shape priors to increase the robustness when dealing with out-of-distribution implant shapes. Our direct estimation method outperforms the baselines provided by the organizers, while the model with shape priors shows superior performance when dealing with out-of-distribution cases. Overall, our methods show promising results in the difficult task of cranial implant design.
Even patients with normal computed tomography (CT) head imaging may experience persistent symptoms for months to years after mild traumatic brain injury (mTBI). There is currently no good way to ...predict recovery and triage patients who may benefit from early follow-up and targeted intervention. We aimed to assess if existing prognostic models can be improved by serum biomarkers or diffusion tensor imaging metrics (DTI) from MRI, and if serum biomarkers can identify patients for DTI.
We included 1025 patients aged >18 years with a Glasgow Coma Score >12 and normal CT from the Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) study which recruited between December 19,2014 and December 17, 2017 (NCT02210221). Biomarkers (GFAP, NFL, S100B) were obtained at a median of 8.8 h (Q1–Q3 4.2–16.7) and DTI at 13 days (3–19) after injury. DTI metrics were available in 153 patients for 48 white matter tracts (ICBM-DTI-81 atlas). Incomplete recovery at three months was defined as an extended Glasgow Outcome Scale score <8. Existing prognostic models were fitted with and without biomarkers, or with and without DTI, and internally validated using bootstrapping.
385 (38%) patients had incomplete recovery. Adding biomarkers did not improve performance beyond the best existing clinical prognostic model optimism-corrected AUC 0.69 (95% CI 0.65–0.72) and R2 17% (11–22). Adding DTI metrics significantly enhanced all models best optimism-corrected AUC 0.82 (0.79–0.85) and R2 75% (39–100). The top three prognostic tracts were the left posterior thalamic radiation, left superior cerebellar peduncle and right uncinate fasciculus. Serum biomarkers could have avoided 1 in 5 DTI scans, with GFAP <12 h and NFL 12–24 h from injury performing best.
DTI substantially improved existing prognostic models for functional outcome in patients with mTBI and a normal CT, and biomarkers could help select patients for MRI. If validated, DTI could allow for targeted follow-up and enrichment of clinical trials of early interventions to improve outcome.
EU Seventh Framework Programme, Hannelore Kohl Stiftung, One Mind, Integra LifeSciences, NeuroTrauma Sciences.
Decompressive craniectomy (DC) is a common surgical procedure consisting of the removal of a portion of the skull that is performed after incidents such as stroke, traumatic brain injury (TBI) or ...other events that could result in acute subdural hemorrhage and/or increasing intracranial pressure. In these cases, CT scans are obtained to diagnose and assess injuries, or guide a certain therapy and intervention. We propose a deep learning based method to reconstruct the skull defect removed during DC performed after TBI from post-operative CT images. This reconstruction is useful in multiple scenarios, e.g. to support the creation of cranioplasty plates, accurate measurements of bone flap volume and total intracranial volume, important for studies that aim to relate later atrophy to patient outcome. We propose and compare alternative self-supervised methods where an encoder-decoder convolutional neural network (CNN) estimates the missing bone flap on post-operative CTs. The self-supervised learning strategy only requires images with complete skulls and avoids the need for annotated DC images. For evaluation, we employ real and simulated images with DC, comparing the results with other state-of-the-art approaches. The experiments show that the proposed model outperforms current manual methods, enabling reconstruction even in highly challenging cases where big skull defects have been removed during surgery.