Deep learning-assisted multiple organ segmentation from whole-body CT images

Abstract

Background: Automated organ segmentation from computed tomography (CT) images facilitates a number of clinical applications, including clinical diagnosis, monitoring of treatment response, quantification, radiation therapy treatment planning, and radiation dosimetry. Purpose: To develop a novel deep learning framework to generate multi-organ masks from CT images for 23 different body organs. Methods: A dataset consisting of 3106 CT images (649,398 axial 2D CT slices, 13,640 images/segment pairs) and ground-truth manual segmentation from various online available databases were collected. After cropping them to body contour, they were resized, normalized and used to train separate models for 23 organs. Data were split to train (80%) and test (20%) covering all the databases. A Res-UNET model was trained to generate segmentation masks from the input normalized CT images. The model output was converted back to the original dimensions and compared with ground-truth segmentation masks in terms of Dice and Jaccard coefficients. The information about organ positions was implemented during post-processing by providing six anchor organ segmentations as input. Our model was compared with the online available TotalSegmentator model through testing our model on their test datasets and their model on our test datasets. Results: The average Dice coefficient before and after post-processing was 84.28% and 83.26% respectively. The average Jaccard index was 76.17 and 70.60 before and after post-processing respectively. Dice coefficients over 90% were achieved for the liver, heart, bones, kidneys, spleen, femur heads, lungs, aorta, eyes, and brain segmentation masks. Post-processing improved the performance in only nine organs. Our model on the TotalSegmentator dataset was better than their models on our dataset in five organs out of 15 common organs and achieved almost similar performance for two organs. Conclusions: The availability of a fast and reliable multi-organ segmentation tool leverages implementation in clinical setting. In this study, we developed deep learning models to segment multiple body organs and compared the performance of our models with different algorithms. Our model was trained on images presenting with large variability emanating from different databases producing acceptable results even in cases with unusual anatomies and pathologies, such as splenomegaly. We recommend using these algorithms for organs providing good performance. One of the main merits of our proposed models is their lightweight nature with an average inference time of 1.67 seconds per case per organ for a total-body CT image, which facilitates their implementation on standard computers.

Fully Automated Explainable Abdominal CT Contrast Media Phase Classification Using Organ Segmentation and Machine Learning

“…The authors have almost followed the same steps, except that they used 48 radiomics features from each organ to predict the contrast media phase and used TotalSegmentator [33] trained models to generate organs masks. The number of features was lower in our study (five simple features), and we used our own segmentation model [23], which is much faster than TotalSegmentor [33] (ten seconds vs three minutes). In terms of accuracy, our model outperformed their reported results.…”

“…These classes for each 3D image were used as the ground truth for training and evaluation of our methodology. Meanwhile, seven organs including the liver, spleen, heart, kidneys, lungs, urinary bladder, and aorta plus body contour were segmented from CT images automatically using previously developed algorithm [23]. Then, five first-order features, including the average, standard deviation (SD), ten percentile, median, and 90 percentiles were extracted using basic Python software-based image processing.…”

“…A previously developed and validated ResUNET-based model was used for CT image segmentation [23]. First, the body contour was generated on CT images via image processing algorithms and then this contour was used to crop the images to foreground.…”

Organomics: A concept reflecting the importance of PET/CT healthy organ radiomics in non-small cell lung cancer prognosis prediction using machine learning

“…We used extended and upgraded versions of previously trained deep learning-based segmentation models in our department [15] to segment 28 volumes of interest in healthy organs on the CT images. Those models were trained using nnU-Net [18] segmentation pipeline using five-fold data split and ensembling all five-folds inferred on the RadioGenomics CT compartment of PET/CT dataset.…”

“…Besides we hypothesize that it may contain some information reflecting overall patients’ health in the radiomic features space from structural (CT) and metabolic (PET) images acquired from these regions. Deep learning-based segmentation enables fast and reliable delimitation of healthy organs and hence evaluation any organ separately [15; 16]. To the best of our knowledge, the contribution of healthy organs is always overlooked and studies exploring the importance of healthy organs to estimate overall patient characteristics in survival prediction in NSCLC patients are lacking.…”