Abdominal Computed Tomography Dataset Research Articles

MedYOLO: A Medical Image Object Detection Framework.

Artificial intelligence-enhanced identification of organs, lesions, and other structures in medical imaging is typically done using convolutional neural networks (CNNs) designed to make voxel-accurate segmentations of the region of interest. However, the labels required to train these CNNs are time-consuming to generate and require attention from subject matter experts to ensure quality. For tasks where voxel-level precision is not required, object detection models offer a viable alternative that can reduce annotation effort. Despite this potential application, there are few options for general-purpose object detection frameworks available for 3-D medical imaging. We report on MedYOLO, a 3-D object detection framework using the one-shot detection method of the YOLO family of models and designed for use with medical imaging. We tested this model on four different datasets: BRaTS, LIDC, an abdominal organ Computed tomography (CT) dataset, and an ECG-gated heart CT dataset. We found our models achieve high performance on a diverse range of structures even without hyperparameter tuning, reaching mean average precision (mAP) at intersection over union (IoU) 0.5 of 0.861 on BRaTS, 0.715 on the abdominal CT dataset, and 0.995 on the heart CT dataset. However, the models struggle with some structures, failing to converge on LIDC resulting in a mAP@0.5 of 0.0.

Wasserstein-GAN-based noise reduction for preserving anatomical structures in low-dose CT

The purpose of low-dose computed tomography (LDCT) is to reduce patient's radiation dose. However, LDCT imaging often suffers from low image quality due to computed tomography (CT) noise patterns. While conventional methods such as iterative reconstruction and image processing have been studied to address this issue, their clinical application remains limited due to the degradation of critical anatomical structures. Recently, deep-learning-based noise reduction approaches have shown promising results in improving the image quality of LDCT. Particularly, generative adversarial network (GAN)-based noise reduction methods have produced outstanding results, approaching the quality of normal-dose CT (NDCT) images. In this study, we propose a hybrid framework based on Wasserstein-GAN with the non-subsampled contourlet transform for enhanced noise reduction in LDCT. The aim is to meticulously address CT noise patterns and generate CT images similar to NDCT. We implemented the proposed algorithm and conducted an experiment using clinical abdominal CT datasets to demonstrate its feasibility. According to our results, the proposed method effectively reduced noise components in LDCT images while preserving the anatomical structures of NDCT. This was confirmed through both qualitative and quantitative evaluations. The implications of our findings for clinical practice are significant, as the proposed method can improve the accuracy of LDCT and reduce radiation dose to patients, ultimately enhancing patient safety.

Recurrent feature propagation and edge skip-connections for automatic abdominal organ segmentation

Automatic segmentation of abdominal organs in computed tomography (CT) images can support radiation therapy and image-guided surgery workflows. Development of such automatic solutions remains challenging mainly owing to complex organ interactions and blurry boundaries in CT images. To address these issues, we focus on effective spatial context modeling and explicit edge segmentation priors. Accordingly, we propose a 3D network with four main components trained end-to-end including shared encoder, edge detector, decoder with edge skip-connections (ESCs) and recurrent feature propagation head (RFP-Head). To capture wide-range spatial dependencies, the RFP-Head propagates and harvests local features through directed acyclic graphs (DAGs) formulated with recurrent connections in an efficient slice-wise manner, with regard to spatial arrangement of image units. To leverage edge information, the edge detector learns edge prior knowledge specifically tuned for semantic segmentation by exploiting intermediate features from the encoder with the edge supervision. The ESCs then aggregate the edge knowledge with multi-level decoder features to learn a hierarchy of discriminative features explicitly modeling complementarity between organs’ interiors and edges for segmentation. We conduct extensive experiments on challenging abdominal CT datasets with eight annotated organs. Experimental results show that the proposed network outperforms several state-of-the-art models, especially for the segmentation of small and complicated structures (gallbladder, esophagus, stomach, pancreas and duodenum).

Noise aware content-noise complementary GAN with local and global discrimination for low-dose CT denoising

In response to rising concerns over radiation exposure in computed tomography (CT) imaging, effective denoising methods for low-dose CT (LDCT) images are crucial. In recent years, the use of deep learning techniques especially generative adversarial networks (GANs) significantly enhanced the efficiency of LDCT denoising methods, surpassing traditional methods. However, GAN-based denoising methods often face challenges in preserving structural consistency and fine details. This study introduces a novel GAN framework with three accretions to enhance the effectiveness of LDCT denoising. Firstly, our generator is designed to leverage a complementary learning scheme between image noise and image content via two distinct paths. One path focuses on exploring the anatomical information of the image, while the second path is dedicated to learning the noise pattern. This complementary learning scheme provides stable noise cancellation while preserving the maximum structural information of the image. Subsequently, we propose a novel noise-conscious mean absolute error loss to address the challenge posed by the non-stationary characteristic of CT noise. In contrast to conventional MAE loss, this loss attentively prioritizes the different parts of the image based on the local distribution of noise in that region. We also incorporate a gradient-domain loss into the loss function, which inspires the generator to preserve precise image details through explicit guidance. Finally, this study adopted a U-Net-based design for the discriminator to better regularize the model by discriminating between the clean image and the denoised image at both global and local levels. The merit of this discriminator is that it can better adapt to the non-stationary environment of GAN training and guide the generator to produce denoised images that are locally and globally consistent. Our thorough experiments using abdominal CT and lung CT datasets demonstrate the superior performance of our approach compared to state-of-the-art methods.

Efficient Multi-Organ Segmentation From 3D Abdominal CT Images With Lightweight Network and Knowledge Distillation.

Accurate segmentation of multiple abdominal organs from Computed Tomography (CT) images plays an important role in computer-aided diagnosis, treatment planning and follow-up. Currently, 3D Convolution Neural Networks (CNN) have achieved promising performance for automatic medical image segmentation tasks. However, most existing 3D CNNs have a large set of parameters and huge floating point operations (FLOPs), and 3D CT volumes have a large size, leading to high computational cost, which limits their clinical application. To tackle this issue, we propose a novel framework based on lightweight network and Knowledge Distillation (KD) for delineating multiple organs from 3D CT volumes. We first propose a novel lightweight medical image segmentation network named LCOV-Net for reducing the model size and then introduce two knowledge distillation modules (i.e., Class-Affinity KD and Multi-Scale KD) to effectively distill the knowledge from a heavy-weight teacher model to improve LCOV-Net's segmentation accuracy. Experiments on two public abdominal CT datasets for multiple organ segmentation showed that: 1) Our LCOV-Net outperformed existing lightweight 3D segmentation models in both computational cost and accuracy; 2) The proposed KD strategy effectively improved the performance of the lightweight network, and it outperformed existing KD methods; 3) Combining the proposed LCOV-Net and KD strategy, our framework achieved better performance than the state-of-the-art 3D nnU-Net with only one-fifth parameters. The code is available at https://github.com/HiLab-git/LCOVNet-and-KD.

Deep Segmentation Networks for Segmenting Kidneys and Detecting Kidney Stones in Unenhanced Abdominal CT Images.

Recent breakthroughs of deep learning algorithms in medical imaging, automated detection, and segmentation techniques for renal (kidney) in abdominal computed tomography (CT) images have been limited. Radiomics and machine learning analyses of renal diseases rely on the automatic segmentation of kidneys in CT images. Inspired by this, our primary aim is to utilize deep semantic segmentation learning models with a proposed training scheme to achieve precise and accurate segmentation outcomes. Moreover, this work aims to provide the community with an open-source, unenhanced abdominal CT dataset for training and testing the deep learning segmentation networks to segment kidneys and detect kidney stones. Five variations of deep segmentation networks are trained and tested both dependently (based on the proposed training scheme) and independently. Upon comparison, the models trained with the proposed training scheme enable the highly accurate 2D and 3D segmentation of kidneys and kidney stones. We believe this work is a fundamental step toward AI-driven diagnostic strategies, which can be an essential component of personalized patient care and improved decision-making in treating kidney diseases.

Deep Encoder-Decoder Adversarial Reconstruction(DEAR) Network for 3D CT from Few-View Data.

X-ray computed tomography (CT) is widely used in clinical practice. The involved ionizing X-ray radiation, however, could increase cancer risk. Hence, the reduction of the radiation dose has been an important topic in recent years. Few-view CT image reconstruction is one of the main ways to minimize radiation dose and potentially allow a stationary CT architecture. In this paper, we propose a deep encoder-decoder adversarial reconstruction (DEAR) network for 3D CT image reconstruction from few-view data. Since the artifacts caused by few-view reconstruction appear in 3D instead of 2D geometry, a 3D deep network has a great potential for improving the image quality in a data driven fashion. More specifically, our proposed DEAR-3D network aims at reconstructing 3D volume directly from clinical 3D spiral cone-beam image data. DEAR is validated on a publicly available abdominal CT dataset prepared and authorized by Mayo Clinic. Compared with other 2D deep learning methods, the proposed DEAR-3D network can utilize 3D information to produce promising reconstruction results.

An Improved Confidence Connected Liver Segmentation Method Based on Three Views of CT Images

To enhance the accuracy of liver segmentation, we present an improved confidence connected liver segmentation method, which combines the liver segmentation results obtained from three views is proposed. First, to reduce noise, an improved curvature anisotropic diffusion filter is applied, which simultaneously stores edge information. Second, seed points located in the liver are selected automatically using statistics and analysis of the liver intensity. We extract the liver contours from three views of computed tomography (CT) images using the confidence connected method and improve the contours by the cavity filling method. Finally, we combine the liver contours extracted from the coronal, sagittal, and cross section. In our experiments, clinical validation is performed using ten abdominal CT datasets. The results show that our proposed method can extract liver contours quickly and accurately, achieving an overall true positive rate (TPR) of 0.97. In addition, this method is useful for clinical diagnosis of liver disorders and virtual surgical planning.

Computed tomography volume measurements of the liver using a liver segmentation and analysis package: an intra- and inter-rater reliability study

Introduction: Recent advances in technology have seen the introduction of software assisted image post-processing (SAIP)tools to calculate the volume of the liver from Computed Tomography (CT) images. One such SAIP tool is the semi-automatedPhilips liver segmentation and analysis package (Endhoven, The Netherlands). The intra- and inter-rater reliability of the livervolumes of 16 participants calculated using this tool was assessed. Methods: Two CT Technologists accessed Abdominal CT data sets and calculated the volume of the liver using the Philips liversegmentation and analysis package. One technologist repeated their calculation. Results: Results demonstrated high intra-rater reliability (ICC 0.9997) and high inter-rater reliability (ICC 0.99994). Analysiswith Bland Altman plots showed the mean bias for both intra- and inter-rater reliability to be close to zero with acceptable limitsof agreement. Conclusions: This study has shown that liver volume measurements using the Philips SAIP (Endhoven, The Netherlands) can betaken reliably by CT technologists, with confidence that the same results would be obtained between different CT technologists.

Optimization of abdominal fat quantification on CT imaging through use of standardized anatomic space: a novel approach.

The quantification of body fat plays an important role in the study of numerous diseases. It is common current practice to use the fat area at a single abdominal computed tomography (CT) slice as a marker of the body fat content in studying various disease processes. This paper sets out to answer three questions related to this issue which have not been addressed in the literature. At what single anatomic slice location do the areas of subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) estimated from the slice correlate maximally with the corresponding fat volume measures? How does one ensure that the slices used for correlation calculation from different subjects are at the same anatomic location? Are there combinations of multiple slices (not necessarily contiguous) whose area sum correlates better with volume than does single slice area with volume? The authors propose a novel strategy for mapping slice locations to a standardized anatomic space so that same anatomic slice locations are identified in different subjects. The authors then study the volume-to-area correlations and determine where they become maximal. To address the third issue, the authors carry out similar correlation studies by utilizing two and three slices for calculating area sum. Based on 50 abdominal CT data sets, the proposed mapping achieves significantly improved consistency of anatomic localization compared to current practice. Maximum correlations are achieved at different anatomic locations for SAT and VAT which are both different from the L4-L5 junction commonly utilized currently for single slice area estimation as a marker. The maximum area-to-volume correlation achieved is quite high, suggesting that it may be reasonable to estimate body fat by measuring the area of fat from a single anatomic slice at the site of maximum correlation and use this as a marker. The site of maximum correlation is not at L4-L5 as commonly assumed, but is more superiorly located at T12-L1 for SAT and at L3-L4 for VAT. Furthermore, the optimal anatomic locations for SAT and VAT estimation are not the same, contrary to common assumption. The proposed standardized space mapping achieves high consistency of anatomic localization by accurately managing nonlinearities in the relationships among landmarks. Multiple slices achieve greater improvement in correlation for VAT than for SAT. The optimal locations in the case of multiple slices are not contiguous.

Blood vessel‐based liver segmentation using the portal phase of an abdominal CT dataset

Blood vessel (BV) information can be used to guide body organ segmentation on computed tomography (CT) imaging. The proposed method uses abdominal BVs (ABVs) to segment the liver through the portal phase of an abdominal CT dataset. This method aims to address the wide variability in liver shape and size, separate liver from other organs of similar intensity, and segment hepatic low-intensity tumors (LITs). Thin ABVs are enhanced using three-dimensional (3D) opening. ABVs are extracted and classified into hepatic BVs (HBVs) and nonhepatic BVs (non-HBVs) with a small number of interactions, and HBVs and non-HBVs are used for constraining automatic liver segmentation. HBVs are used to individually segment the core region of the liver. To separate the liver from other organs, this core region and non-HBVs are used to construct an initial 3D boundary surface. To segment LITs, the core region is classified into non-LIT- and LIT-parts by fitting the histogram of the core region using a variational Bayesian Gaussian mixture model. Each part of the core region is extended based on its corresponding component of the mixture, and extension is completed when it reaches a variation in intensity or the constructed boundary surface, which is reconfirmed to fit robustly between the liver and neighboring organs of similar intensity. A solid-angle technique is used to refine main BVs at the entrances to the inferior vena cava and the portal vein. The proposed method was applied to 80 datasets: 30 Medical Image Computing and Computer Assisted Intervention (MICCAI) and 50 non-MICCAI; 30 datasets of non-MICCAI data include tumors. Our results for MICCAI-test data were evaluated by sliver07 (http://www.sliver07.org/) organizers with an overall score of 85.7, which ranks best on the site as of July 2013. These results (average ± standard deviation) include the five error measures of the 2007 MICCAI workshop for liver segmentation as follows. Results for volume overlap error, relative volume difference, average symmetric surface distance, root mean square symmetric surface distance, and maximum symmetric surface distance were 4.33 ± 0.73, 0.28 ± 0.87, 0.63 ± 0.16, 1.19 ± 0.28, and 14.01 ± 2.88, respectively; and when applying our method to non-MICCAI data, results were 3.21 ± 0.75, 0.06 ± 1.29, 0.45 ± 0.17, 0.98 ± 0.26, and 12.69 ± 3.89, respectively. These results demonstrate high performance of the method when applied to different CT datasets. BVs can be used to address the wide variability in liver shape and size, as BVs provide unique details for the structure of each studied liver. Constructing a boundary surface using HBVs and non-HBVs can separate liver from its neighboring organs of similar intensity. By fitting the histogram of the core region using a variational Bayesian Gaussian mixture model, LITs are segmented and measuring the volumetry of non-LIT- and LIT-parts becomes possible. Further examination of the proposed method on a large number of datasets is required for clinical applications, and development of the method for full automation may be possible and useful in the clinic.

Multi-detector CT in peritoneal carcinomatosis: diagnostic role of thin slices and multiplanar reconstructions

In order to investigate whether 1-mm thin slices and multiplanar reconstructions (MPRs) of multi-detector computed tomography (CT) datasets interpreted in addition to isotropic 5-mm thick slices in one session improve the detection of peritoneal carcinomatosis. The abdominal CT datasets of 44 patients with histologically proven tumors of the abdomen or pelvis were retrospectively evaluated for peritoneal carcinomatosis by four radiologists with variable experience (radiologist 1: >or=10 years, radiologists 2 and 3: 1.5 years, radiologist 4: 0.5 years). In three successive steps, the radiologists evaluated first the axial 5-mm slices, second the 1-mm slices, and third the MPRs and rated their diagnostic confidence. Specificity was nearly unchanged for all the four radiologists. Sensitivity improved for the most experienced and the least experienced radiologists and was unchanged for the two readers with intermediate skills. Except for the third step of radiologist 4, no statistically significant differences in diagnostic performance were detected. The diagnostic confidence of all the four readers benefited to variable degrees from interpretation of the 1-mm slices and MPRs. While 5-mm slices are sufficient for the detection of peritoneal carcinomatosis, 1-mm slices and MPRs can improve sensitivity and diagnostic confidence.

Multidetector-row computed tomography with three-dimensional volume rendering of pancreatic cancer: a complete preoperative staging tool using computed tomography angiography and volume-rendered cholangiopancreatography.

Volume rendering, a postprocessing computer algorithm that creates three-dimensional (3D) displays from computed tomography (CT) datasets, can create 3D cholangiographic images (volume-rendered cholangiopancreatography, or VRCP) from intravenous contrast-enhanced abdominal CT datasets without the use of a biliary contrast agent. This article illustrates the utility of VRCP in the setting of biliary obstruction due to pancreatic cancer. The 3D renderings of the intra- and extrahepatic biliary tree provide valuable information for planning biliary drainage, including the location and length of the obstruction as well as the relationship of intrahepatic ducts to liver metastases.