DeepLesionBrain: Towards a broader deep-learning generalization for multiple sclerosis lesion segmentation

Recently, segmentation methods based on Convolutional Neural Networks (CNNs) showed promising performance in automatic Multiple Sclerosis (MS) lesions segmentation. These techniques have even outperformed human experts in controlled evaluation conditions such as Longitudinal MS Lesion Segmentation Challenge (ISBI Challenge). However, state-of-the-art approaches trained to perform well on highly-controlled datasets fail to generalize on clinical data from unseen datasets. Instead of proposing another improvement of the segmentation accuracy, we propose a novel method robust to domain shift and performing well on unseen datasets, called DeepLesionBrain (DLB). This generalization property results from three main contributions. First, DLB is based on a large group of compact 3D CNNs. This spatially distributed strategy aims to produce a robust prediction despite the risk of generalization failure of some individual networks. Second, we propose a hierarchical specialization learning (HSL) by pre-training a generic network over the whole brain, before using its weights as initialization to locally specialized networks. By this end, DLB learns both generic features extracted at global image level and specific features extracted at local image level. Finally, DLB includes a new image quality data augmentation to reduce dependency to training data specificity (e.g., acquisition protocol). DLB generalization was validated in cross-dataset experiments on MSSEG’16, ISBI challenge, and in-house datasets. During experiments, DLB showed higher segmentation accuracy, better segmentation consistency and greater generalization performance compared to state-of-the-art methods. Therefore, DLB offers a robust framework well-suited for clinical practice.     

Correlation between MRI appearances and serum alkaline phosphatase levels in Osteogenic sarcoma

IntroductionMagnetic resonance imaging is a pictorial depiction of the patho-anatomy of a bony lesion. As different parts of the tumor in Osteogenic sarcoma concurrently undergo various biological processes i.e. osteoblastic new bone formation, cell death, necrosis, bony destruction and revascularization etc., we hypothesized that the image seen in MRI could be used to assess the tumor behavior at that time. This study was done as a preliminary study with the aim to find whether the MRI pictures can have identifiable patterns and if present whether they can be linked to biological behavior. We could identify 2 distinct patterns in T2 weighted images which correlated well with serum alkaline phosphatase a serum marker and the duration of symptoms and so we are reporting our observations.Material and methodT2 weighted MRI pictures of 15 cases of Osteogenic sarcoma were studied for identifiable patterns in matrix. These patterns if found were to be linked to biological behavior in the form of serum alkaline phosphatase levels and duration of symptoms.ResultsWe could identify 2 unique patterns named by us as Group 1 Heterogeneous type (4 cases) which had a raised serum alkaline phosphatase level and had a history at presentation of less than 3 months duration. In Group 2 homogenous type (5 cases) the serum alkaline phosphatase levels were low and the cases presented after 6 months. As we could not identify any logical pattern in rest of the cases we labeled them as miscellaneous.DiscussionMRI patterns can be used as markers of disease activity as there are 2 clear poles correlating well with serum alkaline phosphatase levels (high or low). Intermediate patterns may be the natural biological behavior and waxing and waning of the tumor disease activity.     

Markerless motion capture can provide reliable 3D gait kinematics in the sagittal and frontal plane

Estimating 3D joint rotations in the lower extremities accurately and reliably remains unresolved in markerless motion capture, despite extensive studies in the past decades. The main problems have been ascribed to the limited accuracy of the 3D reconstructions. Accordingly, the purpose of the present study was to develop a new approach based on highly detailed 3D reconstructions in combination with a translational and rotational unconstrained articulated model. The highly detailed 3D reconstructions were synthesized from an eight camera setup using a stereo vision approach. The subject specific articulated model was generated with three rotational and three translational degrees of freedom for each limb segment and without any constraints to the range of motion. This approach was tested on 3D gait analysis and compared to a marker based method. The experiment included ten healthy subjects in whom hip, knee and ankle joint were analysed. Flexion/extension angles as well as hip abduction/adduction closely resembled those obtained from the marker based system. However, the internal/external rotations, knee abduction/adduction and ankle inversion/eversion were less reliable.     

Predictors of diabetes foot complications among patients with diabetes in Saudi Arabia

AimsTo identify risk factors and clinical biomarkers of prevalent diabetes foot complications, including foot ulcers, gangrene and amputations among patients with diabetes in Jeddah, Saudi Arabia.Methods598 diabetes patients from Jeddah participated in the current study. Patients were considered to have diabetes foot complications if they reported diagnosis of foot ulcers or gangrene or amputations in a questionnaire administered by a physician and confirmed by clinical exams. Information on socio-demographic and lifestyle variables was self-reported by patients, and several clinical markers were assessed following standard procedures.ResultsThe prevalence of diabetes foot complications in this population was 11.4%. In the multivariable model without adjustment for PAD (peripheral artery disease) and DPN (diabetes peripheral neuropathy), non-Saudi nationality, longer diabetes duration and insulin use was significantly associated with higher diabetes foot complications prevalence. Each 1 g/L increase of hemoglobin was associated with 2.8% lower prevalence of diabetes foot complications. In the multivariable model adjusting for PAD and DPN, the previously observed associations except for nationality were no longer significant. Patients with both DPN and PAD had 9.73 times the odds of diabetes foot complications compared to the patients with neither condition.ConclusionIn this population, longer diabetes duration, insulin use, lower hemoglobin levels and non-Saudi nationality were associated with higher prevalence of foot complications. These associations were largely explained by the presence of DPN and PAD except for non-Saudi nationality. Diabetes patients with both DPN and PAD had nearly 10-fold increased risk of foot complications than those with neither condition.     

A blockchain based patient centric electronic health record storage and integrity management for e-Health systems

ObjectivesIn the digitization of health records, patients and doctors find difficulty in accessing their health records, especially when the health records are fragmented with different health service providers. In this paper, Patient Centric blockchain smart contracts are designed to provide a regulated solution to the requirements of patients, doctors and health service providers with integrity management.MethodsThe proposed system launches smart contracts namely i) Registration contract for immutable patient log creation. ii) Health Record Creation Contact to generate digital health records. iii) Health Record Storage Contract for secure storage and rapid access with a new Modified Merkle Tree data structure. iv) Update permission Contract that can provide access at emergency situations. v) Data sharing Permission Contract for exchange of health records between different stakeholders and vi) Viewership Permission Contract for viewing the health information by the patients for home care and future care.ResultsThe work has been carried out on a number of trials to check the effectiveness of the proposed system. The qualitative and quantitative metrics of the proposed system have been measured to evaluate the performance of resources, transactions per second, and the latency of transaction.ConclusionThe proposed system provides high security and integrity through cryptographic hash functions. The results are encouraging. The designed system is implemented and found to outperform the existing systems.