Rosmarinic acid ameliorated oxidative stress, neuronal injuries, and mitochondrial dysfunctions mediated by polyglutamine and ɑ-synuclein in Caenorhabditis elegans models.

Numerous natural antioxidants have been developed into agents for neurodegenerative diseases (NDs) treatment. Rosmarinic acid (RA), an excellent antioxidant, exhibits neuroprotective activity, but its anti-NDs efficacy remains puzzling. Here, Caenorhabditis elegans models were employed to systematically reveal RA-mediated mechanisms in delaying NDs from diverse facets, including oxidative stress, the homeostasis of neural and protein, and mitochondrial disorders. Firstly, RA significantly inhibited reactive oxygen species accumulation, reduced peroxide malonaldehyde production, and strengthened the antioxidant defense system via increasing superoxide dismutase activity. Besides, RA reduced neuronal loss and ameliorated polyglutamine and ɑ-synuclein-mediated dyskinesia in NDs models. Further, in combination with the data and molecular docking results, RA may bind specifically to Huntington protein and ɑ-synuclein to prevent toxic protein aggregation and thus enhance proteostasis. Finally, RA ameliorated mitochondrial dysfunction including increasing adenosine triphosphate and mitochondrial membrane potential levels and rescuing mitochondrial membrane proteins' expressions and mitochondrial structural abnormalities via regulating mitochondrial dynamics genes and improving the mitochondrial kinetic homeostasis. Thus, this study systematically revealed the RA-mediated neuroprotective mechanism and promoted RA as a promising nutritional intervention strategy to prevent NDs.

Prediction of Ki-67 expression in bladder cancer based on CT radiomics nomogram.

Objectives: This study aimed to create and validate a radiomics nomogram for non-invasive preoperative Ki-67 expression level prediction in patients with bladder cancer (BCa) using contrast-enhanced CT radiomics features. Methods: A retrospective analysis of 135 patients was conducted, 79 of whom had high levels of Ki-67 expression and 56 of whom had low levels. For the dimensionality reduction analysis, the best features were chosen using the least absolute shrinkage selection operator and one-way analysis of variance. Then, a radiomics nomogram was created using multiple logistic regression analysis based on radiomics features and clinical independent risk factors. The performance of the model was assessed using the Akaike information criterion (AIC) value, the area under the curve (AUC) value, accuracy, sensitivity, and specificity. The clinical usefulness of the model was assessed using decision curve analysis (DCA). Results: Finally, to establish a radiomics nomogram, the best 5 features were chosen and integrated with the independent clinical risk factors (T stage) and Rad-score. This radiomics nomogram demonstrated significant correction and discriminating performance in both the training and validation sets, with an AUC of 0.836 and 0.887, respectively. This radiomics nomogram had the lowest AIC value (AIC = 103.16), which was considered to be the best model. When compared to clinical factor model and radiomics signature, DCA demonstrated the more value of the radiomics nomogram. Conclusion: Enhanced CT-based radiomics nomogram can better predict Ki-67 expression in BCa patients and can be used for prognosis assessment and clinical decision making.

Spatial chromatin accessibility sequencing resolves high-order spatial interactions of epigenomic markers.

As the genome is organized into a three-dimensional structure in intracellular space, epigenomic information also has a complex spatial arrangement. However, most epigenetic studies describe locations of methylation marks, chromatin accessibility regions, and histone modifications in the horizontal dimension. Proper spatial epigenomic information has rarely been obtained. In this study, we designed spatial chromatin accessibility sequencing (SCA-seq) to resolve the genome conformation by capturing the epigenetic information in single-molecular resolution while simultaneously resolving the genome conformation. Using SCA-seq, we are able to examine the spatial interaction of chromatin accessibility (e.g. enhancer-promoter contacts), CpG island methylation, and spatial insulating functions of the CCCTC-binding factor. We demonstrate that SCA-seq paves the way to explore the mechanism of epigenetic interactions and extends our knowledge in 3D packaging of DNA in the nucleus.

CartiMorph: A framework for automated knee articular cartilage morphometrics.

We introduce CartiMorph, a framework for automated knee articular cartilage morphometrics. It takes an image as input and generates quantitative metrics for cartilage subregions, including the percentage of full-thickness cartilage loss (FCL), mean thickness, surface area, and volume. CartiMorph leverages the power of deep learning models for hierarchical image feature representation. Deep learning models were trained and validated for tissue segmentation, template construction, and template-to-image registration. We established methods for surface-normal-based cartilage thickness mapping, FCL estimation, and rule-based cartilage parcellation. Our cartilage thickness map showed less error in thin and peripheral regions. We evaluated the effectiveness of the adopted segmentation model by comparing the quantitative metrics obtained from model segmentation and those from manual segmentation. The root-mean-squared deviation of the FCL measurements was less than 8%, and strong correlations were observed for the mean thickness (Pearson's correlation coefficient ρ∈[0.82,0.97]), surface area (ρ∈[0.82,0.98]) and volume (ρ∈[0.89,0.98]) measurements. We compared our FCL measurements with those from a previous study and found that our measurements deviated less from the ground truths. We observed superior performance of the proposed rule-based cartilage parcellation method compared with the atlas-based approach. CartiMorph has the potential to promote imaging biomarkers discovery for knee osteoarthritis.

Uncertainty-weighted Multi-tasking for T1ρ and T2 Mapping in the Liver with Self-supervised Learning.

Multi-parametric mapping of MRI relaxations in liver has the potential of revealing pathological information of the liver. A self-supervised learning based multi-parametric mapping method is proposed to map T1ρ and T2 simultaneously, by utilising the relaxation constraint in the learning process. Data noise of different mapping tasks is utilised to make the model uncertainty-aware, which adaptively weight different mapping tasks during learning. The method was examined on a dataset of 51 patients with non-alcoholic fatter liver disease. Results showed that the proposed method can produce comparable parametric maps to the traditional multi-contrast pixel wise fitting method, with a reduced number of images and less computation time. The uncertainty weighting also improves the model performance. It has the potential of accelerating MRI quantitative imaging.

Unsupervised domain adaptation for automated knee osteoarthritis phenotype classification.

Background: Osteoarthritis (OA) is a global healthcare problem. The increasing population of OA patients demands a greater bandwidth of imaging and diagnostics. It is important to provide automatic and objective diagnostic techniques to address this challenge. This study demonstrates the utility of unsupervised domain adaptation (UDA) for automated OA phenotype classification. Methods: We collected 318 and 960 three-dimensional double-echo steady-state magnetic resonance images from the Osteoarthritis Initiative (OAI) dataset as the source dataset for phenotype cartilage/meniscus and subchondral bone, respectively. Fifty three-dimensional turbo spin echo (TSE)/fast spin echo (FSE) MR images from our institute were collected as the target datasets. For each patient, the degree of knee OA was initially graded according to the MRI Knee Osteoarthritis Knee Score before being converted to binary OA phenotype labels. The proposed four-step UDA pipeline included (I) pre-processing, which involved automatic segmentation and region-of-interest cropping; (II) source classifier training, which involved pre-training a convolutional neural network (CNN) encoder for phenotype classification using the source dataset; (III) target encoder adaptation, which involved unsupervised adjustment of the source encoder to the target encoder using both the source and target datasets; and (IV) target classifier validation, which involved statistical analysis of the classification performance evaluated by the area under the receiver operating characteristic curve (AUROC), sensitivity, specificity and accuracy. We compared our model on the target data with the source pre-trained model and the model trained with the target data from scratch. Results: For phenotype cartilage/meniscus, our model has the best performance out of the three models, giving 0.90 [95% confidence interval (CI): 0.79-1.02] of the AUROC score, while the other two model show 0.52 (95% CI: 0.13-0.90) and 0.76 (95% CI: 0.53-0.98). For phenotype subchondral bone, our model gave 0.75 (95% CI: 0.56-0.94) at AUROC, which has a close performance of the source pre-trained model (0.76, 95% CI: 0.55-0.98), and better than the model trained from scratch on the target dataset only (0.53, 95% CI: 0.33-0.73). Conclusions: By utilising a large, high-quality source dataset for training, the proposed UDA approach enhances the performance of automated OA phenotype classification for small target datasets. As a result, our technique enables improved downstream analysis of locally collected datasets with a small sample size.

Improving the RNA velocity approach with single-cell RNA lifecycle (nascent, mature and degrading RNAs) sequencing technologies.

We presented an experimental method called FLOUR-seq, which combines BD Rhapsody and nanopore sequencing to detect the RNA lifecycle (including nascent, mature, and degrading RNAs) in cells. Additionally, we updated our HIT-scISOseq V2 to discover a more accurate RNA lifecycle using 10x Chromium and Pacbio sequencing. Most importantly, to explore how single-cell full-length RNA sequencing technologies could help improve the RNA velocity approach, we introduced a new algorithm called 'Region Velocity' to more accurately configure cellular RNA velocity. We applied this algorithm to study spermiogenesis and compared the performance of FLOUR-seq with Pacbio-based HIT-scISOseq V2. Our findings demonstrated that 'Region Velocity' is more suitable for analyzing single-cell full-length RNA data than traditional RNA velocity approaches. These novel methods could be useful for researchers looking to discover full-length RNAs in single cells and comprehensively monitor RNA lifecycle in cells.

An uncertainty aided framework for learning based liverT1ρmapping and analysis.

Objective. QuantitativeT1ρimaging has potential for assessment of biochemical alterations of liver pathologies. Deep learning methods have been employed to accelerate quantitativeT1ρimaging. To employ artificial intelligence-based quantitative imaging methods in complicated clinical environment, it is valuable to estimate the uncertainty of the predicatedT1ρvalues to provide the confidence level of the quantification results. The uncertainty should also be utilized to aid the post-hoc quantitative analysis and model learning tasks.Approach. To address this need, we propose a parametric map refinement approach for learning-basedT1ρmapping and train the model in a probabilistic way to model the uncertainty. We also propose to utilize the uncertainty map to spatially weight the training of an improvedT1ρmapping network to further improve the mapping performance and to remove pixels with unreliableT1ρvalues in the region of interest. The framework was tested on a dataset of 51 patients with different liver fibrosis stages.Main results. Our results indicate that the learning-based map refinement method leads to a relative mapping error of less than 3% and provides uncertainty estimation simultaneously. The estimated uncertainty reflects the actual error level, and it can be used to further reduce relativeT1ρmapping error to 2.60% as well as removing unreliable pixels in the region of interest effectively.Significance. Our studies demonstrate the proposed approach has potential to provide a learning-based quantitative MRI system for trustworthyT1ρmapping of the liver.

Repeatability of quantitative T1rho magnetic resonance imaging in normal brain tissues at 3.0T.

PURPOSE: T1rho imaging is a promising MRI technique for imaging of brain disease. This study aimed to assess the repeatability of quantitative T1rho imaging in the normal brain grey and white matter. METHODS: The study prospectively recruited 30 healthy volunteers without a history of neurological diseases or brain injury, and T1rho was performed and quantified from three imaging sessions. Repeat measures analysis of variance (ANOVA) and within-subject coefficients of variation (wCoV) was used to detect differences in T1rho values between the three scans. RESULTS: The results showed low wCoVs of less than 4.3% (range 0.92-4.27%) across all the brain structures. No significant differences were observed in T1rho measurement between the three scans (p > 0.05). The amygdala and hippocampus showed the highest T1rho values of 91.79 ± 2.55 msec and 91.07 ± 2.11 msec respectively, and the palladium and putamen had the lowest values of 67.60 ± 1.84 msec and 71.83 ± 1.85 msec respectively. CONCLUSION: T1rho shows high test-retest repeatability for whole brain imaging in serial imaging sessions, indicating it to be a reliable sequence for quantitative brain imaging.

BIND&MODIFY: a long-range method for single-molecule mapping of chromatin modifications in eukaryotes.

Epigenetic modifications of histones are associated with development and pathogenesis of disease. Existing approaches cannot provide insights into long-range interactions and represent the average chromatin state. Here we describe BIND&MODIFY, a method using long-read sequencing for profiling histone modifications and transcription factors on individual DNA fibers. We use recombinant fused protein A-M.EcoGII to tether methyltransferase M.EcoGII to protein binding sites to label neighboring regions by methylation. Aggregated BIND&MODIFY signal matches bulk ChIP-seq and CUT&TAG. BIND&MODIFY can simultaneously measure histone modification status, transcription factor binding, and CpG 5mC methylation at single-molecule resolution and also quantifies correlation between local and distal elements.

A CT-based radiomics nomogram for differentiation of benign and malignant small renal masses (≤4 cm).

RATIONALE AND OBJECTIVES: Based on radiomics signature and clinical data, to develop and verify a radiomics nomogram for preoperative distinguish between benign and malignant of small renal masses (SRM). MATERIALS AND METHODS: One hundred and fifty-six patients with malignant (n = 92) and benign (n = 64) SRM were divided into the following three categories: category A, typical angiomyolipoma (AML) with visible fat; category B, benign SRM without visible fat, including fat-poor angiomyolipoma (fp-AML), and other rare benign renal tumors; category C, malignant renal tumors. At the same time, one hundred and fifty-six patients included in the study were divided into the training set (n = 108) and test set (n = 48). Respectively from corticomedullary phase (CP), nephrogram phase (NP) and excretory phase (EP) CT images to extract the radiomics features, and the optimal features were screened to establish the logistic regression model and decision tree model, and computed the radiomics score (Rad-score). Demographics and CT findings were evaluated and statistically significant factors were selected to construct a clinical factors model. The radiomics nomogram was established by merging Rad-score and selected clinical factors. The Akaike information criterion (AIC) values and the area under the curve (AUC) were used to compare model discriminant performance, and decision curve analysis (DCA) was used to assess clinical usefulness. RESULTS: Seven, fifteen, nineteen, and seventeen distinguishing features were obtained in the CP, NP, EP, and three-phase joint, respectively, and the logistic regression and decision tree models were built based on this features. In the training set, the logistic regression model works better than the decision tree model for distinguishing categories A and B from category C, with the AUC of CP, NP, EP and three-phase joint were 0.868, 0.906, 0.937 and 0.975, respectively. The radiomics nomogram constructed based on the three-phase joint Rad-score and selected clinical factor performed well on the training set (AUC, 0.988; 95% CI, 0.974-1.000) for differentiation of categories A and B from category C. In the test set, the AUC of clinical factors model, radiomics signature and radiomics nomogram for discriminating categories A and B from category C were 0.814, 0.954 and 0.968, respectively; for the identification of category A from category C, the AUC of the three models were 0.789, 0.979, 0.985, respectively; for discriminating category B from category C, the AUC of the three models were 0.853, 0.915, 0.946, respectively. The radiomics nomogram had better discriminative than the clinical factors model in both training and test sets (P < 0.05). The radiomics nomogram (AIC = 40.222) with the lowest AIC value was considered the best model compared with that of the clinical factors model (AIC = 106.814) and the radiomics signature (AIC = 44.224). The DCA showed that the radiomics nomogram have better clinical utility than the clinical factors model and radiomics signature. CONCLUSIONS: The logistic regression model has better discriminative performance than the decision tree model, and the radiomics nomogram based on Rad-score of three-phase joint and clinical factors has a good predictive effect in differentiating benign from malignant of SRM, which may help clinicians develop accurate and individualized treatment strategies.

Detecting Early-Stage Liver Fibrosis Using Macromolecular Proton Fraction Mapping Based on Spin-Lock MRI: Preliminary Observations.

BACKGROUND: Liver fibrosis is characterized by macromolecule depositions. Recently, a novel technology termed macromolecular proton fraction quantification based on spin-lock magnetic resonance imaging (MPF-SL) is reported to measure macromolecule levels. HYPOTHESIS: MPF-SL can detect early-stage liver fibrosis by measuring macromolecule levels in the liver. STUDY TYPE: Retrospective. SUBJECTS: Fifty-five participants, including 22 with no fibrosis (F0) and 33 with early-stage fibrosis (F1-2), were recruited. FIELD STRENGTH/SEQUENCE: 3 T; two-dimensional (2D) MPF-SL turbo spin-echo sequence, 2D spin-lock T1rho turbo spin-echo sequence, and multi-slice 2D gradient echo sequence. ASSESSMENT: Macromolecular proton fraction (MPF), T1rho, liver iron concentration (LIC), and fat fraction (FF) biomarkers were quantified within regions of interest. STATISTICAL TESTS: Group comparison of the biomarkers using Mann-Whitney U tests; correlation between the biomarkers assessed using Spearman's rank correlation coefficient and linear regression with goodness-of-fit; fibrosis stage differentiation using receiver operating characteristic curve (ROC) analysis. P-value < 0.05 was considered statistically significant. RESULTS: Average T1rho was 41.76 ± 2.94 msec for F0 and 41.15 ± 3.73 msec for F1-2 (P = 0.60). T1rho showed nonsignificant correlation with either liver fibrosis (ρ = -0.07; P = 0.61) or FF (ρ = -0.14; P = 0.35) but indicated a negative correlation with LIC (ρ = -0.66). MPF was 4.73 ± 0.45% and 5.65 ± 0.81% for F0 and F1-2 participants, respectively. MPF showed a positive correlation with liver fibrosis (ρ = 0.59), and no significant correlations with LIC (ρ = 0.02; P = 0.89) or FF (ρ = 0.05; P = 0.72). The area under the ROC curve was 0.85 (95% confidence interval [CI] 0.75-0.95) and 0.55 (95% CI 0.39-0.71; P = 0.55) for MPF and T1rho to discriminate between F0 and F1-2 fibrosis, respectively. DATA CONCLUSION: MPF-SL has the potential to diagnose early-stage liver fibrosis and does not appear to be confounded by either LIC or FF. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 3.

CT-based deep learning radiomics signature for the preoperative prediction of the muscle-invasive status of bladder cancer.

Objectives: Although the preoperative assessment of whether a bladder cancer (BCa) indicates muscular invasion is crucial for adequate treatment, there currently exist some challenges involved in preoperative diagnosis of BCa with muscular invasion. The aim of this study was to construct deep learning radiomic signature (DLRS) for preoperative predicting the muscle invasion status of BCa. Methods: A retrospective review covering 173 patients revealed 43 with pathologically proven muscle-invasive bladder cancer (MIBC) and 130 with non-muscle-invasive bladder cancer (non- MIBC). A total of 129 patients were randomly assigned to the training cohort and 44 to the test cohort. The Pearson correlation coefficient combined with the least absolute shrinkage and selection operator (LASSO) was utilized to reduce radiomic redundancy. To decrease the dimension of deep learning features, Principal Component Analysis (PCA) was adopted. Six machine learning classifiers were finally constructed based on deep learning radiomics features, which were adopted to predict the muscle invasion status of bladder cancer. The area under the curve (AUC), accuracy, sensitivity and specificity were used to evaluate the performance of the model. Results: According to the comparison, DLRS-based models performed the best in predicting muscle violation status, with MLP (Train AUC: 0.973260 (95% CI 0.9488-0.9978) and Test AUC: 0.884298 (95% CI 0.7831-0.9855)) outperforming the other models. In the test cohort, the sensitivity, specificity and accuracy of the MLP model were 0.91 (95% CI 0.551-0.873), 0.78 (95% CI 0.594-0.863) and 0.58 (95% CI 0.729-0.827), respectively. DCA indicated that the MLP model showed better clinical utility than Radiomics-only model, which was demonstrated by the decision curve analysis. Conclusions: A deep radiomics model constructed with CT images can accurately predict the muscle invasion status of bladder cancer.

Denoising of three-dimensional fast spin echo magnetic resonance images of knee joints using spatial-variant noise-relevant residual learning of convolution neural network.

PURPOSE: Two-dimensional (2D) fast spin echo (FSE) techniques play a central role in the clinical magnetic resonance imaging (MRI) of knee joints. Moreover, three-dimensional (3D) FSE provides high-isotropic-resolution magnetic resonance (MR) images of knee joints, but it has a reduced signal-to-noise ratio compared to 2D FSE. Deep-learning denoising methods are a promising approach for denoising MR images, but they are often trained using synthetic noise due to challenges in obtaining true noise distributions for MR images. In this study, inherent true noise information from two number of excitations (2-NEX) acquisition was used to develop a deep-learning model based on residual learning of convolutional neural network (CNN), and this model was used to suppress the noise in 3D FSE MR images of knee joints. METHODS: A deep learning-based denoising method was developed. The proposed CNN used two-step residual learning over parallel transporting and residual blocks and was designed to comprehensively learn real noise features from 2-NEX training data. RESULTS: The results of an ablation study validated the network design. The new method achieved improved denoising performance of 3D FSE knee MR images compared with current state-of-the-art methods, based on the peak signal-to-noise ratio and structural similarity index measure. The improved image quality after denoising using the new method was verified by radiological evaluation. CONCLUSION: A deep CNN using the inherent spatial-varying noise information in 2-NEX acquisitions was developed. This method showed promise for clinical MRI assessments of the knee, and has potential applications for the assessment of other anatomical structures.

Uncertainty-aware self-supervised neural network for liverT1ρmapping with relaxation constraint.

Objective.T1ρmapping is a promising quantitative MRI technique for the non-invasive assessment of tissue properties. Learning-based approaches can mapT1ρfrom a reduced number ofT1ρweighted images but requires significant amounts of high-quality training data. Moreover, existing methods do not provide the confidence level of theT1ρestimation. We aim to develop a learning-based liverT1ρmapping approach that can mapT1ρwith a reduced number of images and provide uncertainty estimation.Approach. We proposed a self-supervised neural network that learns aT1ρmapping using the relaxation constraint in the learning process. Epistemic uncertainty and aleatoric uncertainty are modelled for theT1ρquantification network to provide a Bayesian confidence estimation of theT1ρmapping. The uncertainty estimation can also regularize the model to prevent it from learning imperfect data. Main results. We conducted experiments onT1ρdata collected from 52 patients with non-alcoholic fatty liver disease. The results showed that when only collecting twoT1ρ-weighted images, our method outperformed the existing methods forT1ρquantification of the liver. Our uncertainty estimation can further regularize the model to improve the performance of the model and it is consistent with the confidence level of liverT1ρvalues.Significance. Our method demonstrates the potential for accelerating theT1ρmapping of the liver by using a reduced number of images. It simultaneously provides uncertainty ofT1ρquantification which is desirable in clinical applications.

Dual-Ligand Near-Infrared Luminescent Lanthanide-Based Metal-Organic Framework Coupled with In Vivo Microdialysis for Highly Sensitive Ratiometric Detection of Zn2+ in a Mouse Model of Alzheimer's Disease.

Zinc, which is the second most abundant trace element in the human central nervous system, is closely associated with Alzheimer's disease (AD). However, attempts to develop highly sensitive and selective sensing systems for Zn2+ in the brain have not been successful. Here, we used a one-step solvothermal method to design and prepare a metal-organic framework (MOF) containing the dual ligands, terephthalic acid (H2BDC) and 2,2':6',2″-terpyridine (TPY), with Eu3+ as a metal node. This MOF is denoted as Eu-MOF/BDC-TPY. Adjustment of the size and morphology of Eu-MOF/BDC-TPY allowed the dual ligands to produce multiple luminescence peaks, which could be interpreted via ratiometric fluorescence to detect Zn2+ using the ratio of Eu3+-based emission, as the internal reference, and ligand-based emission, as the indicator. Thus, Eu-MOF/BDC-TPY not only displayed higher selectivity than other metal cations but also offered a highly accurate, sensitive, wide linear, color change-based technique for detecting Zn2+ at concentrations ranging from 1 nM to 2 µM, with a low limit of detection (0.08 nM). Moreover, Eu-MOF/BDC-TPY maintained structural stability and displayed a fluorescence intensity of at least 95.4% following storage in water for 6 months. More importantly, Eu-MOF/BDC-TPY sensed the presence of Zn2+ markedly rapidly (within 5 s), which was very useful in practical application. Furthermore, the results of our ratiometric luminescent method-based analysis of Zn2+ in AD mouse brains were consistent with those obtained using inductively coupled plasma mass spectrometry.

Deep collaborative network with alpha matte for precise knee tissue segmentation from MRI.

BACKGROUND AND OBJECTIVE: Precise segmentation of knee tissues from magnetic resonance imaging (MRI) is critical in quantitative imaging and diagnosis. Convolutional neural networks (CNNs), being state of the art, often challenged by the lack of image-specific adaptation, such as low tissue contrasts and structural inhomogeneities, thereby leading to incomplete segmentation results. METHODS: This paper presents a deep learning-based automatic segmentation framework for precise knee tissue segmentation. A novel deep collaborative method is proposed, which consists of an encoder-decoder-based segmentation network in combination with a low rank tensor-reconstructed segmentation network. Low rank reconstruction in MRI tensor sub-blocks is introduced to exploit the morphological variations in knee tissues. To model the tissue boundary regions and effectively utilize the superimposed regions, trimap generation is proposed for defining high, medium and low confidence regions from the multipath CNNs. The secondary path with low rank reconstructed input mitigates the conditions in which the primary segmentation network can potentially fail and overlook the boundary regions. The outcome of the segmentation is solved as an alpha matting problem by blending the trimap with the source input. RESULTS: Experiments on Osteoarthritis Initiative (OAI) datasets with all the 6 musculoskeletal tissues provide an overall segmentation dice score of 0.8925, where Femoral and Tibial part of cartilage achieving an average dice exceeding 0.9. The volumetric metrics also indicate the superior performances in all tissue compartments. CONCLUSIONS: Experiments on Osteoarthritis Initiative (OAI) datasets and a self-prepared scan validate the effectiveness of the proposed method. Inclusion of extra prediction scale allowed the model to distinguish and segment the tissue boundary accurately. We specifically demonstrate the application of the proposed method in a cartilage segmentation-based thickness map for diagnosis purposes.