MRI lung lobe segmentation in pediatric cystic fibrosis patients using a recurrent neural network trained with publicly accessible CT datasets.
Ontology highlight
ABSTRACT: Purpose
To introduce a widely applicable workflow for pulmonary lobe segmentation of MR images using a recurrent neural network (RNN) trained with chest CT datasets. The feasibility is demonstrated for 2D coronal ultrafast balanced SSFP (ufSSFP) MRI.Methods
Lung lobes of 250 publicly accessible CT datasets of adults were segmented with an open-source CT-specific algorithm. To match 2D ufSSFP MRI data of pediatric patients, both CT data and segmentations were translated into pseudo-MR images that were masked to suppress anatomy outside the lung. Network-1 was trained with pseudo-MR images and lobe segmentations and then applied to 1000 masked ufSSFP images to predict lobe segmentations. These outputs were directly used as targets to train Network-2 and Network-3 with non-masked ufSSFP data as inputs, as well as an additional whole-lung mask as input for Network-2. Network predictions were compared to reference manual lobe segmentations of ufSSFP data in 20 pediatric cystic fibrosis patients. Manual lobe segmentations were performed by splitting available whole-lung segmentations into lobes.Results
Network-1 was able to segment the lobes of ufSSFP images, and Network-2 and Network-3 further increased segmentation accuracy and robustness. The average all-lobe Dice similarity coefficients were 95.0 ± 2.8 (mean ± pooled SD [%]) and 96.4 ± 2.5, 93.0 ± 2.0; and the average median Hausdorff distances were 6.1 ± 0.9 (mean ± SD [mm]), 5.3 ± 1.1, 7.1 ± 1.3 for Network-1, Network-2, and Network-3, respectively.Conclusion
Recurrent neural network lung lobe segmentation of 2D ufSSFP imaging is feasible, in good agreement with manual segmentations. The proposed workflow might provide access to automated lobe segmentations for various lung MRI examinations and quantitative analyses.
SUBMITTER: Pusterla O
PROVIDER: S-EPMC9314108 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:ObjectivesDensitometry on paired inspiratory and expiratory multidetector computed tomography (MDCT) for the quantification of air trapping is an important approach to assess functional changes in airways diseases such as cystic fibrosis (CF). For a regional analysis of functional deficits, an accurate lobe segmentation algorithm applicable to inspiratory and expiratory scans is beneficial.Materials and methodsWe developed a fully automated lobe segmentation algorithm, and subsequently validated automatically generated lobe masks (ALM) against manually corrected lobe masks (MLM). Paired inspiratory and expiratory CTs from 16 children with CF (mean age 11.1±2.4) acquired at 4 time-points (baseline, 3mon, 12mon, 24mon) with 2 kernels (B30f, B60f) were segmented, resulting in 256 ALM. After manual correction spatial overlap (Dice index) and mean differences in lung volume and air trapping were calculated for ALM vs. MLM.ResultsThe mean overlap calculated with Dice index between ALM and MLM was 0.98±0.02 on inspiratory, and 0.86±0.07 on expiratory CT. If 6 lobes were segmented (lingula treated as separate lobe), the mean overlap was 0.97±0.02 on inspiratory, and 0.83±0.08 on expiratory CT. The mean differences in lobar volumes calculated in accordance with the approach of Bland and Altman were generally low, ranging on inspiratory CT from 5.7±52.23cm3 for the right upper lobe to 17.41±14.92cm3 for the right lower lobe. Higher differences were noted on expiratory CT. The mean differences for air trapping were even lower, ranging from 0±0.01 for the right upper lobe to 0.03±0.03 for the left lower lobe.ConclusionsAutomatic lobe segmentation delivers excellent results for inspiratory and good results for expiratory CT. It may become an important component for lobe-based quantification of functional deficits in cystic fibrosis lung disease, reducing necessity for user-interaction in CT post-processing.
| S-EPMC5890971 | biostudies-literature
Project description:The enormous increase in the amount of publicly available chemical genomics data and the growing emphasis on data sharing and open science mandates that cheminformaticians also make their models publicly available for broad use by the scientific community. Chembench is one of the first publicly accessible, integrated cheminformatics Web portals. It has been extensively used by researchers from different fields for curation, visualization, analysis, and modeling of chemogenomics data. Since its launch in 2008, Chembench has been accessed more than 1 million times by more than 5000 users from a total of 98 countries. We report on the recent updates and improvements that increase the simplicity of use, computational efficiency, accuracy, and accessibility of a broad range of tools and services for computer-assisted drug design and computational toxicology available on Chembench. Chembench remains freely accessible at https://chembench.mml.unc.edu.
| S-EPMC5720369 | biostudies-literature
Project description:BackgroundThis paper considers automatic segmentation of the left cardiac ventricle in short axis magnetic resonance images. Various aspects, such as the presence of papillary muscles near the endocardium border, makes simple threshold based segmentation difficult.MethodsThe endo- and epicardium are modelled as two series of radii which are inter-related using features describing shape and motion. Image features are derived from edge information from human annotated images. The features are combined within a discriminatively trained Conditional Random Field (CRF). Loopy belief propagation is used to infer segmentations when an unsegmented video sequence is given. Powell's method is applied to find CRF parameters by minimizing the difference between ground truth annotations and the inferred contours. We also describe how the endocardium centre points are calculated from a single human-provided centre point in the first frame, through minimization of frame alignment error.ResultsWe present and analyse the results of segmentation. The algorithm exhibits robustness against inclusion of the papillary muscles by integrating shape and motion information. Possible future improvements are identified.ConclusionsThe presented model integrates shape and motion information to segment the inner and outer contours in the presence of papillary muscles. On the Sunnybrook dataset we find an average Dice metric of 0.91 ± 0.02 and 0.93 ± 0.02 for the inner and outer segmentations, respectively. Particularly problematic are patients with hypertrophy where the blood pool disappears from view at end-systole.
| S-EPMC3737053 | biostudies-literature
Project description:BackgroundBody composition (BC) is an important factor in determining the risk of type 2-diabetes and cardiovascular disease. Computed tomography (CT) is a useful imaging technique for studying BC, however manual segmentation of CT images is time-consuming and subjective. The purpose of this study is to develop and evaluate fully automated segmentation techniques applicable to a 3-slice CT imaging protocol, consisting of single slices at the level of the liver, abdomen, and thigh, allowing detailed analysis of numerous tissues and organs.MethodsThe study used more than 4000 CT subjects acquired from the large-scale SCAPIS and IGT cohort to train and evaluate four convolutional neural network based architectures: ResUNET, UNET++, Ghost-UNET, and the proposed Ghost-UNET++. The segmentation techniques were developed and evaluated for automated segmentation of the liver, spleen, skeletal muscle, bone marrow, cortical bone, and various adipose tissue depots, including visceral (VAT), intraperitoneal (IPAT), retroperitoneal (RPAT), subcutaneous (SAT), deep (DSAT), and superficial SAT (SSAT), as well as intermuscular adipose tissue (IMAT). The models were trained and validated for each target using tenfold cross-validation and test sets.ResultsThe Dice scores on cross validation in SCAPIS were: ResUNET 0.964 (0.909-0.996), UNET++ 0.981 (0.927-0.996), Ghost-UNET 0.961 (0.904-0.991), and Ghost-UNET++ 0.968 (0.910-0.994). All four models showed relatively strong results, however UNET++ had the best performance overall. Ghost-UNET++ performed competitively compared to UNET++ and showed a more computationally efficient approach.ConclusionFully automated segmentation techniques can be successfully applied to a 3-slice CT imaging protocol to analyze multiple tissues and organs related to BC. The overall best performance was achieved by UNET++, against which Ghost-UNET++ showed competitive results based on a more computationally efficient approach. The use of fully automated segmentation methods can reduce analysis time and provide objective results in large-scale studies of BC.
| S-EPMC10506248 | biostudies-literature
Project description:Recent advances in MRI and increasing knowledge on the characterization and anatomical variability of medial temporal lobe (MTL) anatomy have paved the way for more specific subdivisions of the MTL in humans. In addition, recent studies suggest that early changes in many neurodegenerative and neuropsychiatric diseases are better detected in smaller subregions of the MTL rather than with whole structure analyses. Here, we developed a new protocol using 7 Tesla (T) MRI incorporating novel anatomical findings for the manual segmentation of entorhinal cortex (ErC), perirhinal cortex (PrC; divided into area 35 and 36), parahippocampal cortex (PhC), and hippocampus; which includes the subfields subiculum (Sub), CA1, CA2, as well as CA3 and dentate gyrus (DG) which are separated by the endfolial pathway covering most of the long axis of the hippocampus. We provide detailed instructions alongside slice-by-slice segmentations to ease learning for the untrained but also more experienced raters. Twenty-two subjects were scanned (19-32 yrs, mean age = 26 years, 12 females) with a turbo spin echo (TSE) T2-weighted MRI sequence with high-resolution oblique coronal slices oriented orthogonal to the long axis of the hippocampus (in-plane resolution 0.44 × 0.44 mm2) and 1.0 mm slice thickness. The scans were manually delineated by two experienced raters, to assess intra- and inter-rater reliability. The Dice Similarity Index (DSI) was above 0.78 for all regions and the Intraclass Correlation Coefficients (ICC) were between 0.76 to 0.99 both for intra- and inter-rater reliability. In conclusion, this study presents a fine-grained and comprehensive segmentation protocol for MTL structures at 7 T MRI that closely follows recent knowledge from anatomical studies. More specific subdivisions (e.g. area 35 and 36 in PrC, and the separation of DG and CA3) may pave the way for more precise delineations thereby enabling the detection of early volumetric changes in dementia and neuropsychiatric diseases.
| S-EPMC5476466 | biostudies-literature
Project description:IntroductionMicro-computed tomography (μCT) is a valuable imaging modality for longitudinal quantification of bone volumes to identify disease or treatment effects for a broad range of conditions that affect bone health. Complex structures, such as the hindpaw with up to 31 distinct bones in mice, have considerable analytic potential, but quantification is often limited to a single bone volume metric due to the intensive effort of manual segmentation. Herein, we introduce a high-throughput, user-friendly, and semi-automated method for segmentation of murine hindpaw μCT datasets.MethodsIn vivo μCT was performed on male (n = 4; 2-8-months) and female (n = 4; 2-5-months) C57BL/6 mice longitudinally each month. Additional 9.5-month-old male C57BL/6 hindpaws (n = 6 hindpaws) were imaged by ex vivo μCT to investigate the effects of resolution and integration time on analysis outcomes. The DICOMs were exported to Amira software for the watershed-based segmentation, and watershed markers were generated automatically at approximately 80% accuracy before user correction. The semi-automated segmentation method utilizes the original data, binary mask, and bone-specific markers that expand to the full volume of the bone using watershed algorithms.ResultsCompared to the conventional manual segmentation using Scanco software, the semi-automated approach produced similar raw bone volumes. The semi-automated segmentation also demonstrated a significant reduction in segmentation time for both experienced and novice users compared to standard manual segmentation. ICCs between experienced and novice users were >0.9 (excellent reliability) for all but 4 bones.DiscussionThe described semi-automated segmentation approach provides remarkable reliability and throughput advantages. Adoption of the semi-automated segmentation approach will provide standardization and reliability of bone volume measures across experienced and novice users and between institutions. The application of this model provides a considerable strategic advantage to accelerate various research opportunities in pre-clinical bone and joint analysis towards clinical translation.
| S-EPMC8816671 | biostudies-literature
Project description:BackgroundAutomated segmentation of spinal magnetic resonance imaging (MRI) plays a vital role both scientifically and clinically. However, accurately delineating posterior spine structures is challenging.MethodsThis retrospective study, approved by the ethical committee, involved translating T1-weighted and T2-weighted images into computed tomography (CT) images in a total of 263 pairs of CT/MR series. Landmark-based registration was performed to align image pairs. We compared two-dimensional (2D) paired - Pix2Pix, denoising diffusion implicit models (DDIM) image mode, DDIM noise mode - and unpaired (SynDiff, contrastive unpaired translation) image-to-image translation using "peak signal-to-noise ratio" as quality measure. A publicly available segmentation network segmented the synthesized CT datasets, and Dice similarity coefficients (DSC) were evaluated on in-house test sets and the "MRSpineSeg Challenge" volumes. The 2D findings were extended to three-dimensional (3D) Pix2Pix and DDIM.Results2D paired methods and SynDiff exhibited similar translation performance and DCS on paired data. DDIM image mode achieved the highest image quality. SynDiff, Pix2Pix, and DDIM image mode demonstrated similar DSC (0.77). For craniocaudal axis rotations, at least two landmarks per vertebra were required for registration. The 3D translation outperformed the 2D approach, resulting in improved DSC (0.80) and anatomically accurate segmentations with higher spatial resolution than that of the original MRI series.ConclusionsTwo landmarks per vertebra registration enabled paired image-to-image translation from MRI to CT and outperformed all unpaired approaches. The 3D techniques provided anatomically correct segmentations, avoiding underprediction of small structures like the spinous process.Relevance statementThis study addresses the unresolved issue of translating spinal MRI to CT, making CT-based tools usable for MRI data. It generates whole spine segmentation, previously unavailable in MRI, a prerequisite for biomechanical modeling and feature extraction for clinical applications.Key points• Unpaired image translation lacks in converting spine MRI to CT effectively. • Paired translation needs registration with two landmarks per vertebra at least. • Paired image-to-image enables segmentation transfer to other domains. • 3D translation enables super resolution from MRI to CT. • 3D translation prevents underprediction of small structures.
| S-EPMC10643734 | biostudies-literature
Project description:BACKGROUND: Cystic lymphangioma is a rare benign lesion derived from the detachment of the lymph sacs from venous drainage systems; the treatment of choice is a surgical excision and the final diagnosis is of histological type. PURPOSE: To compare the results of ultrasound (US), computed tomography (CT), and magnetic resonance imaging (MRI) in patients with cystic lymphangioma to clearly evaluate the anatomic as well as the structural lesion features necessary for differential diagnosis and for the patient treatment planning. MATERIAL AND METHODS: We analyzed the imaging results of six patients admitted in our department to evaluate cyst-like tumor masses clinically palpable or detected by US. All the patients underwent US, CT, and MRI. The pathology reports demonstrated a mesenterial cystic lymphangioma in five cases underwent surgical resection and in the last case a chest cystic lymphangioma underwent a fine needle aspiration biopsy (FNAB). RESULTS: In all the cases, the results of US, CT, and MRI were concordant showing cyst-like tumor masses in the abdomen (n = 5) and chest (n = 1) ranging in size from 3.5 to 15 cm. CONCLUSION: According to our experience, we suggest that the appropriate diagnostic imaging protocol in patients with cystic lymphangioma should initially include the US study and followed by a MRI scan with contrast administration. CT should be avoided because of radiation exposure. US and MRI may also be useful in the follow-up of patients who refuse surgical resection or in whom surgery is contraindicated or postponed as well as to early detect a possible disease relapse.
| S-EPMC4437906 | biostudies-other
Project description:The glioblastoma microenvironment predominantly contains tumor-associated macrophages that support tumor growth and invasion. We investigated the relationship between tumor radiosensitivity and infiltrating M1/M2 macrophage profiles in public datasets of primary and recurrent glioblastoma. We estimated the radiosensitivity index (RSI) score based on gene expression rankings. Macrophages were profiled using the deconvolution algorithm CIBERSORTx. Samples from The Cancer Genome Atlas (TCGA), Chinese Glioma Genome Atlas (CGGA), the Ivy Glioblastoma Atlas Project dataset, a single-cell RNA sequencing dataset (GSE84465), Glioma Longitudinal Analysis Consortium (GLASS), and an immunotherapy trial dataset (GSE121810) were included. RSI-high radioresistant tumors were associated with worse overall survival in TCGA and CGGA than RSI-low tumors. M1/M2 macrophage ratios and RSI scores were inversely associated, indicating that radioresistant glioblastoma tumor microenvironments contain more M2 than M1 macrophages. In the single-cell RNA sequencing dataset, the mean RSI of neoplastic cells was positively correlated with high M2 macrophages proportions. A favorable response to programmed cell death protein 1 (PD-1) therapy was observed in recurrent glioblastomas with high M1/M2 macrophage ratios and low RSI scores. In patients with recurrent glioblastoma, fewer M2 macrophages and low RSI scores were associated with improved overall survival. High M2 macrophage proportions may be involved in radioresistant glioblastoma.
| S-EPMC8869561 | biostudies-literature
Project description:We present an adversarial domain adaptation based deep learning approach for automatic tumor segmentation from T2-weighted MRI. Our approach is composed of two steps: (i) a tumor-aware unsupervised cross-domain adaptation (CT to MRI), followed by (ii) semi-supervised tumor segmentation using Unet trained with synthesized and limited number of original MRIs. We introduced a novel target specific loss, called tumor-aware loss, for unsupervised cross-domain adaptation that helps to preserve tumors on synthesized MRIs produced from CT images. In comparison, state-of-the art adversarial networks trained without our tumor-aware loss produced MRIs with ill-preserved or missing tumors. All networks were trained using labeled CT images from 377 patients with non-small cell lung cancer obtained from the Cancer Imaging Archive and unlabeled T2w MRIs from a completely unrelated cohort of 6 patients with pre-treatment and 36 on-treatment scans. Next, we combined 6 labeled pre-treatment MRI scans with the synthesized MRIs to boost tumor segmentation accuracy through semi-supervised learning. Semi-supervised training of cycle-GAN produced a segmentation accuracy of 0.66 computed using Dice Score Coefficient (DSC). Our method trained with only synthesized MRIs produced an accuracy of 0.74 while the same method trained in semi-supervised setting produced the best accuracy of 0.80 on test. Our results show that tumor-aware adversarial domain adaptation helps to achieve reasonably accurate cancer segmentation from limited MRI data by leveraging large CT datasets.
| S-EPMC6169798 | biostudies-literature