Automatic multiorgan segmentation in CT images of the male pelvis using region-specific hierarchical appearance cluster models

Statistical multi-level shape models for scalable modeling of multi-organ anatomies

Statistical shape modeling is an indispensable tool in the quantitative analysis of anatomies. Particle-based shape modeling (PSM) is a state-of-the-art approach that enables the learning of population-level shape representation from medical imaging data (e.g., CT, MRI) and the associated 3D models of anatomy generated from them. PSM optimizes the placement of a dense set of landmarks (i.e., correspondence points) on a given shape cohort. PSM supports multi-organ modeling as a particular case of the conventional single-organ framework via a global statistical model, where multi-structure anatomy is considered as a single structure. However, global multi-organ models are not scalable for many organs, induce anatomical inconsistencies, and result in entangled shape statistics where modes of shape variation reflect both within- and between-organ variations. Hence, there is a need for an efficient modeling approach that can capture the inter-organ relations (i.e., pose variations) of the complex anatomy while simultaneously optimizing the morphological changes of each organ and capturing the population-level statistics. This paper leverages the PSM approach and proposes a new approach for correspondence-point optimization of multiple organs that overcomes these limitations. The central idea of multilevel component analysis, is that the shape statistics consists of two mutually orthogonal subspaces: the within-organ subspace and the between-organ subspace. We formulate the correspondence optimization objective using this generative model. We evaluate the proposed method using synthetic shape data and clinical data for articulated joint structures of the spine, foot and ankle, and hip joint.

Development and Validation of an Interpretable Artificial Intelligence Model to Predict 10-Year Prostate Cancer Mortality

Simple Summary

This article presents a gradient-boosted model that can predict 10-year prostate cancer mortality with high accuracy. The model was developed and validated on prospective multicenter data from the PLCO trial. Using XGBoost and Shapley values, it provides interpretability to understand its prediction. It can be used online to provide predictions and support informed decision-making in PCa treatment.

Abstract

Prostate cancer treatment strategies are guided by risk-stratification. This stratification can be difficult in some patients with known comorbidities. New models are needed to guide strategies and determine which patients are at risk of prostate cancer mortality. This article presents a gradient-boosting model to predict the risk of prostate cancer mortality within 10 years after a cancer diagnosis, and to provide an interpretable prediction. This work uses prospective data from the PLCO Cancer Screening and selected patients who were diagnosed with prostate cancer. During follow-up, 8776 patients were diagnosed with prostate cancer. The dataset was randomly split into a training (n = 7021) and testing (n = 1755) dataset. Accuracy was 0.98 (±0.01), and the area under the receiver operating characteristic was 0.80 (±0.04). This model can be used to support informed decision-making in prostate cancer treatment. AI interpretability provides a novel understanding of the predictions to the users.

FRFCM clustering segmentation method for medical MR image feature diagnosis

Medical image recognition is affected by characteristics such as blur and noise, which cause medical image features that cannot be effectively identified and directly affects clinical diagnostics. In order to improve the diagnostic effect of medical MR image features, based on the FRFCM clustering segmentation method, this study combines the medical MR image feature reality, collects data for traditional clustering method analysis, and sorts out the shortcomings of traditional clustering methods. Simultaneously, this study improves the traditional clustering method by combining medical image feature diagnosis requirements. In addition, this study carried out image data processing through simulation, and designed comparative experiments to analyze the performance of the algorithm. The research shows that the FRFCM combined with the intuitionistic fuzzy set proposed in this paper has greatly improved the noise immunity and segmentation performance compared with the FCM based fuzzy set.

Auto-segmentations by convolutional neural network in cervical and anorectal cancer with clinical structure sets as the ground truth

•

The network provided auto-segmentations with high overlap with ground truth volumes.

•

Evaluation of femoral heads/bladder auto-segmentations showed highest overlap.

•

Annotated structure sets from daily clinical practice is feasible as ground truth.

Background

It is time-consuming for oncologists to delineate volumes for radiotherapy treatment in computer tomography (CT) images. Automatic delineation based on image processing exists, but with varied accuracy and moderate time savings. Using convolutional neural network (CNN), delineations of volumes are faster and more accurate. We have used CTs with the annotated structure sets to train and evaluate a CNN.

Material and methods

The CNN is a standard segmentation network modified to minimize memory usage. We used CTs and structure sets from 75 cervical cancers and 191 anorectal cancers receiving radiation therapy at Skåne University Hospital 2014-2018. Five structures were investigated: left/right femoral heads, bladder, bowel bag, and clinical target volume of lymph nodes (CTVNs). Dice score and mean surface distance (MSD) (mm) evaluated accuracy, and one oncologist qualitatively evaluated auto-segmentations.

Results

Median Dice/MSD scores for anorectal cancer: 0.91–0.92/1.93–1.86 femoral heads, 0.94/2.07 bladder, and 0.83/6.80 bowel bag. Median Dice scores for cervical cancer were 0.93–0.94/1.42–1.49 femoral heads, 0.84/3.51 bladder, 0.88/5.80 bowel bag, and 0.82/3.89 CTVNs. With qualitative evaluation, performance on femoral heads and bladder auto-segmentations was mostly excellent, but CTVN auto-segmentations were not acceptable to a larger extent.

Discussion

It is possible to train a CNN with high overlap using structure sets as ground truth. Manually delineated pelvic volumes from structure sets do not always strictly follow volume boundaries and are sometimes inaccurately defined, which leads to similar inaccuracies in the CNN output. More data that is consistently annotated is needed to achieve higher CNN accuracy and to enable future clinical implementation.

CT-based multi-organ segmentation using a 3D self-attention U-net network for pancreatic radiotherapy

PURPOSE

Segmentation of organs-at-risk (OARs) is a weak link in radiotherapeutic treatment planning process because the manual contouring action is labor-intensive and time-consuming. This work aimed to develop a deep learning-based method for rapid and accurate pancreatic multi-organ segmentation that can expedite the treatment planning process.

METHODS

We retrospectively investigated one hundred patients with computed tomography (CT) simulation scanned and contours delineated. Eight OARs including large bowel, small bowel, duodenum, left kidney, right kidney, liver, spinal cord and stomach were the target organs to be segmented. The proposed three-dimensional (3D) deep attention U-Net is featured with a deep attention strategy to effectively differentiate multiple organs. Performance of the proposed method was evaluated using six metrics, including Dice similarity coefficient (DSC), sensitivity, specificity, Hausdorff distance 95% (HD95), mean surface distance (MSD) and residual mean square distance (RMSD).

RESULTS

The contours generated by the proposed method closely resemble the ground-truth manual contours, as evidenced by encouraging quantitative results in terms of DSC, sensitivity, specificity, HD95, MSD and RMSD. For DSC, mean values of 0.91 ± 0.03, 0.89 ± 0.06, 0.86 ± 0.06, 0.95 ± 0.02, 0.95 ± 0.02, 0.96 ± 0.01, 0.87 ± 0.05 and 0.93 ± 0.03 were achieved for large bowel, small bowel, duodenum, left kidney, right kidney, liver, spinal cord and stomach, respectively.

CONCLUSIONS

The proposed method could significantly expedite the treatment planning process by rapidly segmenting multiple OARs. The method could potentially be used in pancreatic adaptive radiotherapy to increase dose delivery accuracy and minimize gastrointestinal toxicity.

Fully automated organ segmentation in male pelvic CT images

Accurate segmentation of prostate and surrounding organs at risk is important for prostate cancer radiotherapy treatment planning. We present a fully automated workflow for male pelvic CT image segmentation using deep learning. The architecture consists of a 2D organ volume localization network followed by a 3D segmentation network for volumetric segmentation of prostate, bladder, rectum, and femoral heads. We used a multi-channel 2D U-Net followed by a 3D U-Net with encoding arm modified with aggregated residual networks, known as ResNeXt. The models were trained and tested on a pelvic CT image dataset comprising 136 patients. Test results show that 3D U-Net based segmentation achieves mean (±SD) Dice coefficient values of 90 (±2.0)%, 96 (±3.0)%, 95 (±1.3)%, 95 (±1.5)%, and 84 (±3.7)% for prostate, left femoral head, right femoral head, bladder, and rectum, respectively, using the proposed fully automated segmentation method.

The segmentation of bones in pelvic CT images based on extraction of key frames

Background

Bone segmentation is important in computed tomography (CT) imaging of the pelvis, which assists physicians in the early diagnosis of pelvic injury, in planning operations, and in evaluating the effects of surgical treatment. This study developed a new algorithm for the accurate, fast, and efficient segmentation of the pelvis.

Methods

The proposed method consists of two main parts: the extraction of key frames and the segmentation of pelvic CT images. Key frames were extracted based on pixel difference, mutual information and normalized correlation coefficient. In the pelvis segmentation phase, skeleton extraction from CT images and a marker-based watershed algorithm were combined to segment the pelvis. To meet the requirements of clinical application, physician’s judgment is needed. Therefore the proposed methodology is semi-automated.

Results

In this paper, 5 sets of CT data were used to test the overlapping area, and 15 CT images were used to determine the average deviation distance. The average overlapping area of the 5 sets was greater than 94%, and the minimum average deviation distance was approximately 0.58 pixels. In addition, the key frame extraction efficiency and the running time of the proposed method were evaluated on 20 sets of CT data. For each set, approximately 13% of the images were selected as key frames, and the average processing time was approximately 2 min (the time for manual marking was not included).

Conclusions

The proposed method is able to achieve accurate, fast, and efficient segmentation of pelvic CT image sequences. Segmentation results not only provide an important reference for early diagnosis and decisions regarding surgical procedures, they also offer more accurate data for medical image registration, recognition and 3D reconstruction.

Automatic segmentation of the clinical target volume and organs at risk in the planning CT for rectal cancer using deep dilated convolutional neural networks

PURPOSE

Delineation of the clinical target volume (CTV) and organs at risk (OARs) is very important for radiotherapy but is time-consuming and prone to inter-observer variation. Here, we proposed a novel deep dilated convolutional neural network (DDCNN)-based method for fast and consistent auto-segmentation of these structures.

METHODS

Our DDCNN method was an end-to-end architecture enabling fast training and testing. Specifically, it employed a novel multiple-scale convolutional architecture to extract multiple-scale context features in the early layers, which contain the original information on fine texture and boundaries and which are very useful for accurate auto-segmentation. In addition, it enlarged the receptive fields of dilated convolutions at the end of networks to capture complementary context features. Then, it replaced the fully connected layers with fully convolutional layers to achieve pixel-wise segmentation. We used data from 278 patients with rectal cancer for evaluation. The CTV and OARs were delineated and validated by senior radiation oncologists in the planning computed tomography (CT) images. A total of 218 patients chosen randomly were used for training, and the remaining 60 for validation. The Dice similarity coefficient (DSC) was used to measure segmentation accuracy.

RESULTS

Performance was evaluated on segmentation of the CTV and OARs. In addition, the performance of DDCNN was compared with that of U-Net. The proposed DDCNN method outperformed the U-Net for all segmentations, and the average DSC value of DDCNN was 3.8% higher than that of U-Net. Mean DSC values of DDCNN were 87.7% for the CTV, 93.4% for the bladder, 92.1% for the left femoral head, 92.3% for the right femoral head, 65.3% for the intestine, and 61.8% for the colon. The test time was 45 s per patient for segmentation of all the CTV, bladder, left and right femoral heads, colon, and intestine. We also assessed our approaches and results with those in the literature: our system showed superior performance and faster speed.

CONCLUSIONS

These data suggest that DDCNN can be used to segment the CTV and OARs accurately and efficiently. It was invariant to the body size, body shape, and age of the patients. DDCNN could improve the consistency of contouring and streamline radiotherapy workflows.