Manual Contouring Research Articles

Atlas-based lung segmentation combined with automatic densitometry characterization in COVID-19 patients: Training, validation and first application in a longitudinal study

PurposeTo develop and validate an automated segmentation tool for COVID-19 lung CTs. To combine it with densitometry information in identifying Aerated, Intermediate and Consolidated Volumes in admission (CT1) and follow up CT (CT3). Materials and MethodsAn Atlas was trained on manually segmented CT1 of 250 patients and validated on 10 CT1 of the training group, 10 new CT1 and 10 CT3, by comparing DICE index between automatic (AUTO), automatic-corrected (AUTOMAN) and manual (MAN) contours. A previously developed automatic method was applied on HU lung density histograms to quantify Aerated, Intermediate and Consolidated Volumes. Volumes of subregions in validation CT1 and CT3 were quantified for each method. ResultsIn validation CT1/CT3, manual correction of automatic contours was not necessary in 40% of cases. Mean DICE values for both lungs were 0.94 for AUTOVsMAN and 0.96 for AUTOMANVsMAN. Differences between Aerated and Intermediate Volumes quantified with AUTOVsMAN contours were always < 6%. Consolidated Volumes showed larger differences (mean: −95 ± 72 cc). If considering AUTOMANVsMAN volumes, differences got further smaller for Aerated and Intermediate, and were drastically reduced for consolidated Volumes (mean: −36 ± 25 cc). The average time for manual correction of automatic lungs contours on CT1 was 5 ± 2 min. ConclusionsAn Atlas for automatic segmentation of lungs in COVID-19 patients was developed and validated. Combined with a previously developed method for lung densitometry characterization, it provides a fast, operator-independent way to extract relevant quantitative parameters with minimal manual intervention.

The geometric and dosimetric effect of algorithm choice on propagated contours from CT to cone beam CTs

PurposeAdaptive radiotherapy relies of rapid re-contouring, online more so than offline. Intra-patient contour propagation via non-rigid registration offers a solution but can be of limited accuracy. However, the dosimetric significance of the inaccuracies is unknown. Here we evaluate the dosimetric reliability of contours generated by different commercially-available software packages. MethodPlanning CT contours for ten head and neck cancer patients were propagated via five commercial packages to five CBCT scans acquired throughout treatment. The treatment plan was recalculated on each of the CBCTs for each set of propagated contours, and DVH parameters extracted for the spinal cord, brainstem, parotids and larynx. The propagated contours were compared to two gold standard contours: contours manually outlined and a consensus STAPLE contours generated from the propagated contours. Geometrical similarity was evaluated using mean distance to agreement (mDTA), Hausdorff distance, centroid agreement and Dice similarity coefficient. Dosimetric reliability was assessed against clinical constraints and comparing via the intraclass correlation coefficient (ICC). ResultsAll propagated contours were similar to the STAPLE (mDTA < 1.0 mm) whilst larger differences were seen for the manual contours (mDTA < 3.0 mm). The dosimetric comparison showed that the propagated contours gave excellent dose estimates for most organs. The spinal cord reliability was moderate (ICC > 0.66). ConclusionsLarge differences in geometric metrics rarely had a statistically significant impact on DVH parameters for the OARs studied. For that reason, propagated contours on treatment CBCT images are suitable for estimating dose to the OARs.

Reliability and Validity of an Ultrasound-Based Protocol for Measurement of Quadriceps Muscle Thickness in Children.

Introduction and aims: Accurate determination of skeletal muscle size is of great importance in multiple settings including resistance exercise, aging, disease, and disuse. Ultrasound (US) measurement of muscle thickness (MT) is a method of relatively high availability and low cost. The present study aims to evaluate a multisite ultrasonographic protocol for measurement of MT with respect to reproducibility and correlation to gold-standard measurements of muscle volume (MV) with magnetic resonance imaging (MRI) in children. Material and methods: 15 children completed the study (11 ± 1 year, 41 ± 8 kg, 137 ± 35 cm). Following 20 min supine rest, two investigators performed US MT measurements of all four heads of the m. quadriceps femoris, at pre-determined sites. Subsequently, MRI scanning was performed and MV was estimated by manual contouring of individual muscle heads. Results: Ultrasound measurement of MT had an intra-rater reliability of ICC = 0.985–0.998 (CI 95% = 0.972–0.998) and inter-rater reliability of ICC = 0.868–0.964 (CI 95% = 0.637–0.983). The US examinations took less than 15 min, per investigator. Muscle thickness of all individual quadriceps muscles correlated significantly with their corresponding MV as measured by MRI (overall r = 0.789, p < 0.001). Conclusion: The results of this study indicate that US measurement of MT using a multisite protocol is a competitive alternative to MRI scanning, especially with respect to availability and time consumption. Therefore, US MT could allow for wider clinical and scientific implementation.

CAM-Wnet: An effective solution for accurate pulmonary embolism segmentation.

The morbidity of pulmonary embolism (PE) is only lower than that of coronary heart disease and hypertension. Early detection, early diagnosis, and timely treatment are the keys to effectively reduce the risk of death. Nevertheless, PE segmentation is still a challenging task at present. The automatic segmentation of PE is particularly important. On the one hand, manual segmentation of PE from a computed tomography (CT) sequence is very time-consuming and prone to misdiagnose. On the other hand, an accurate contour of the location, volume, and shape of PE can help radiotherapists carry out targeted treatment and thus greatly increase the survival rate of patients. Therefore, developing an automatic and efficient PE segmentation approach is an urgent demand in clinical diagnosis. An accurate segmentation of PE is critical for the diagnosis of PE. However, it remains a difficult and relevant problem in the field of medical image processing due to factors like incongruent sizes and shapes of emboli regions, and low contrast between embolisms and other tissues. To address this conundrum, in this study, a deep neural network (CAM-Wnet) that incorporates coordinate attention (CA) mechanisms and pyramid pooling modules (PPMs) is proposed to end-to-end segment PE from CT image. CAM-Wnet architecture is composed of coarse U-Net and subdivision U-Net stacked on top of each other. First, the coarse U-Net uses a pretrained VGG-19 as an encoder, which can transfer the features learned from ImageNet to other tasks. At the same time, CA residual blocks (CARBs) are introduced into the decoder of the coarse network to obtain a wider range of semantic information and find out the correlation between channels. Then, the multiplied results of input image and preliminary mask are put into the subdivision U-Net for secondary feature distillation, and the encoder and decoder of the subdivision U-Net are both constructed from CARBs, too. The PPMs are used between the encoder and the decoder of two U-Net architectures to utilize global context information and further enhance the feature extraction effect. Finally, the improved focal loss function is used to train the network to further improve the segmentation effect. In this study, we used the doctors' manual contours of the China-Japan Friendship Hospital dataset to test the proposed architecture. We calculated the Precision, Recall, IoU, and F1-score to evaluate the accuracy of the architecture for PE segmentation. The segmentation Precision for PE was found to be 0.9703, Recall was 0.963, IoU was 0.9353, and F1-score was 0.9665. The experimental results show the effectiveness of the proposed method to automatically and accurately segment embolism in lung CT images. Furthermore, we also test the performance of our method on the liver tumor segmentation public dataset, which demonstrates the effectiveness and generalization ability of our method. CAM-Wnet obtained more global information and semantic information with the introduction of multiscale pooling and attention mechanisms. Experimental results showed that the proposed method effectively improved the segmentation effect of PE in lung CT images and could be applied to assist doctors in clinical treatment.

Histopathological validation of semi-automated myocardial scar quantification techniques for dark-blood late gadolinium enhancement magnetic resonance imaging.

To evaluate the performance of various semi-automated techniques for quantification of myocardial infarct size on both conventional bright-blood and novel dark-blood late gadolinium enhancement (LGE) images using histopathology as reference standard. In 13 Yorkshire pigs, reperfused myocardial infarction was experimentally induced. At 7 weeks post-infarction, both bright-blood and dark-blood LGE imaging were performed on a 1.5 T magnetic resonance scanner. Following magnetic resonance imaging (MRI), the animals were sacrificed, and histopathology was obtained. The percentage of infarcted myocardium was assessed per slice using various semi-automated scar quantification techniques, including the signal threshold vs. reference mean (STRM, using 3 to 8 SDs as threshold) and full-width at half-maximum (FWHM) methods, as well as manual contouring, for both LGE methods. Infarct size obtained by histopathology was used as reference. In total, 24 paired LGE MRI slices and histopathology samples were available for analysis. For both bright-blood and dark-blood LGE, the STRM method with a threshold of 5 SDs led to the best agreement to histopathology without significant bias (-0.23%, 95% CI [-2.99, 2.52%], P = 0.862 and -0.20%, 95% CI [-2.12, 1.72%], P = 0.831, respectively). Manual contouring significantly underestimated infarct size on bright-blood LGE (-1.57%, 95% CI [-2.96, -0.18%], P = 0.029), while manual contouring on dark-blood LGE outperformed semi-automated quantification and demonstrated the most accurate quantification in this study (-0.03%, 95% CI [-0.22, 0.16%], P = 0.760). The signal threshold vs. reference mean method with a threshold of 5 SDs demonstrated the most accurate semi-automated quantification of infarcted myocardium, without significant bias compared to histopathology, for both conventional bright-blood and novel dark-blood LGE.

Automated detection and segmentation of non-small cell lung cancer computed tomography images

Detection and segmentation of abnormalities on medical images is highly important for patient management including diagnosis, radiotherapy, response evaluation, as well as for quantitative image research. We present a fully automated pipeline for the detection and volumetric segmentation of non-small cell lung cancer (NSCLC) developed and validated on 1328 thoracic CT scans from 8 institutions. Along with quantitative performance detailed by image slice thickness, tumor size, image interpretation difficulty, and tumor location, we report an in-silico prospective clinical trial, where we show that the proposed method is faster and more reproducible compared to the experts. Moreover, we demonstrate that on average, radiologists & radiation oncologists preferred automatic segmentations in 56% of the cases. Additionally, we evaluate the prognostic power of the automatic contours by applying RECIST criteria and measuring the tumor volumes. Segmentations by our method stratified patients into low and high survival groups with higher significance compared to those methods based on manual contours.

Differences in tumor-to-normal organ SUV ratios measured with 68 Ga-DOTATATE PET compared with 177 Lu-DOTATATE SPECT in patients with neuroendocrine tumors.

Our goal is to quantitatively compare radiotracer biodistributions within tumors and major normal organs on pretherapy 68 Ga-DOTATATE PET to post-therapy 177 Lu-DOTATATE single-photon emission computed tomography (SPECT) in patients receiving peptide receptor radionuclide therapy (PRRT). PET/CT at ~ 60 min postinjection of Ga-68 DOTATATE and research 177 Lu-SPECT/CT imaging ~ at 4 h (SPECT1) and ~ 24 h (SPECT2) post-cycle#1 were available. Manual contours of lesions on baseline CT or MRI were applied to co-registered SPECT/CT and PET/CT followed by deep learning-based CT auto-segmentation of organs. Tumor-to-normal organ ratios (TNR) were calculated from standardized uptake values (SUV) mean and SUV peak for tumor, and SUV mean for non-tumoral liver (nliver), spleen and kidney. There were 90 lesons in 24 patients with progressive metastatic neuroendocrine tumor. The correlation between PET and SPECT SUV TNRs were poor/moderate: PET versus SPECT1 R 2 = 0.19, 0.21, 0.29; PET versus SPECT2 R 2 = 0.06, 0.16, 0.33 for TNR nliver ,TNR spleen ,TNR kidney , respectively. Across all patients, the average value of the TNR measured on PET was significantly lower than on SPECT at both time points ( P < 0.001). Using SUV mean for tumor, average TNR values and 95% confidence intervals (CI) were PET: TNR nliver = 3.5 [CI: 3.0-3.9], TNR spleen = 1.3 [CI, 1.2-1.5], TNR kidney = 1.7 [CI: 1.6-1.9]; SPECT1: TNR nliver = 10 [CI: 8.2-11.7], TNR spleen = 2.9 [CI: 2.5-3.4], TNR kidney = 2.8 [CI: 2.3-3.3]; SPECT2: TNR nliver = 16.9 [CI: 14-19.9], TNR spleen = 3.6 [CI: 3-4.2], TNR kidney = 3.6 [CI: 3.0-4.2]. Comparison of PET and SPECT results in a sphere phantom study demonstrated that these differences are not attributed to imaging modality. Differences in TNR exist for the theranostic pair, with significantly higher SUV TNR on 177 Lu SPECT compared with 68 Ga PET. We postulate this phenomenon is due to temporal differences in DOTATATE uptake and internalization in tumor as compared to normal organs.

Impact of Using Unedited CT-Based DIR-Propagated Autocontours on Online ART for Pancreatic SBRT.

PurposeTo determine the dosimetric impact of using unedited autocontours in daily plan adaptation of patients with locally advanced pancreatic cancer (LAPC) treated with stereotactic body radiotherapy using tumor tracking.Materials and MethodsThe study included 98 daily CT scans of 35 LAPC patients. All scans were manually contoured (MAN), and included the PTV and main organs-at-risk (OAR): stomach, duodenum and bowel. Precision and MIM deformable image registration (DIR) methods followed by contour propagation were used to generate autocontour sets on the daily CT scans. Autocontours remained unedited, and were compared to MAN on the whole organs and at 3, 1 and 0.5 cm from the PTV. Manual and autocontoured OAR were used to generate daily plans using the VOLO™ optimizer, and were compared to non-adapted plans. Resulting planned doses were compared based on PTV coverage and OAR dose-constraints.ResultsOverall, both algorithms reported a high agreement between unclipped MAN and autocontours, but showed worse results when being evaluated on the clipped structures at 1 cm and 0.5 cm from the PTV. Replanning with unedited autocontours resulted in better OAR sparing than non-adapted plans for 95% and 84% plans optimized using Precision and MIM autocontours, respectively, and obeyed OAR constraints in 64% and 56% of replans.ConclusionFor the majority of fractions, manual correction of autocontours could be avoided or be limited to the region closest to the PTV. This practice could further reduce the overall timings of adaptive radiotherapy workflows for patients with LAPC.

Killer Whale (Orcinus orca) Pulsed Calls in the Eastern Canadian Arctic

&#x0D; Killer whales (Orcinus orca) produce a variety of acoustic signal types used for communication: clicks, whistles, and pulsed calls. Discrete pulsed calls are highly stereotyped, repetitive, and unique to individual pods found around the world. Discriminating amongst pod specific calls can help determine population structure in killer whales and is used to track pod movements around oceans. Killer whale presence in the Canadian Arctic has increased substantially, but we have limited understanding of their ecology, movements, and stock identity. Two autonomous passive acoustic monitoring (PAM) hydrophones were deployed in the waters of Eclipse Sound and Milne Inlet, in northern Baffin Island, Nunavut, Canada, in August and September 2017. Eleven killer whale pulsed call types, three multiphonic and eight monophonic, are proposed and described using manual whistle contour extraction and feature normalization. Automated detection of echolocation clicks between 20 and 48 kHz demonstrated little to no overlap between killer whale calls and echolocation presumed to be narwhal, which suggests that narwhal remain audibly inconspicuous when killer whales are present. Describing the acoustic repertoire of killer whales seasonally present in the Canadian Arctic will aid in understanding their acoustic behaviour, seasonal movements, and ecological impacts. The calls described here provide a basis for future acoustic comparisons across the North Atlantic and aid in characterizing killer whale demographics and ecology, particularly for pods making seasonal incursions into Arctic waters.

Patient specific deep learning based segmentation for magnetic resonance guided prostate radiotherapy.

Background and PurposeTreatments on combined Magnetic Resonance (MR) scanners and Linear Accelerators (Linacs) for radiotherapy, called MR-Linacs, often require daily contouring. Currently, deformable image registration (DIR) algorithms propagate contours from reference scans, however large shape and size changes can be troublesome. Artificial neural network (ANN) based contouring may alleviate this issue, however generally requires large datasets for training. Mitigating the problem of scarcity of data, we propose patient specific networks trained on a single dataset for each patient, for contouring onto the following datasets in an adaptive MR-Linacworkflow.Materials and MethodsMR-scans from 17 prostate patients treated on an MR-Linac with contours of Clinical Target Volume (CTV), bladder and rectum were utilized. U-net shaped models were trained based on the image from the first fraction of each patient, and subsequently applied onto the following treatment images. Results were compared with manual contours in terms of the Dice coefficient and Added Path Length (APL). As benchmark, contours propagated through the clinical DIR algorithm were similarly evaluated.ResultsIn Dice coefficient the ANN output was 0.92 ± 0.03, 0.93 ± 0.07 and 0.84 ± 0.10 while for DIR 0.95 ± 0.03, 0.93 ± 0.08, 0.88 ± 0.06 for CTV, bladder and rectum respectively. Similarly, APL where 3109 ± 1642, 7250 ± 4234 and 5041 ± 2666 for ANN and 1835 ± 1621, 7236 ± 4287 and 4170 ± 2920 voxels for DIR.ConclusionsPatient specific ANN models trained on images from the first fraction of a prostate MR-Linac treatment showed similar accuracy when applied to the subsequent fraction images as a clinically implemented DIR method.

Automatic stent recognition using perceptual attention U-net for quantitative intrafraction motion monitoring in pancreatic cancer radiotherapy.

Stent has often been used as an internal surrogate to monitor intrafraction tumor motion during pancreatic cancer radiotherapy. Based on the stent contours generated from planning CT images, the current intrafraction motion review (IMR) system on Varian TrueBeam only provides a tool to verify the stent motion visually but lacks quantitative information. The purpose of this study is to develop an automatic stent recognition method for quantitative intrafraction tumor motion monitoring in pancreatic cancer treatment. A total of 535 IMR images from 14 pancreatic cancer patients were retrospectively selected in this study, with the manual contour of the stent on each image serving as the ground truth. We developed a deep learning-based approach that integrates two mechanisms that focus on the features of the segmentation target. The objective attention modeling was integrated into the U-net framework to deal with the optimization difficulties when training a deep network with 2D IMR images and limited training data. A perceptual loss was combined with the binary cross-entropy loss and a Dice loss for supervision. The deep neural network was trained to capture more contextual information to predict binary stent masks. A random-split test was performed, with images of ten patients (71%, 380 images) randomly selected for training, whereas the rest of four patients (29%, 155 images) were used for testing. Sevenfold cross-validation of the proposed PAUnet on the 14 patients was performed for further evaluation. Our stent segmentation results were compared with the manually segmented contours. For the random-split test, the trained model achieved a mean (±standard deviation) stent Dice similarity coefficient (DSC), 95% Hausdorff distance (HD95), the center-of-mass distance (CMD), and volume difference were 0.96 (±0.01), 1.01 (±0.55) mm, 0.66 (±0.46) mm, and 3.07% (±2.37%), respectively. The sevenfold cross-validation of the proposed PAUnet had the mean (±standard deviation) of 0.96 (±0.02), 0.72 (±0.49) mm, 0.85 (±0.96) mm, and 3.47% (±3.27%) for the DSC, HD95, CMD, and . We developed a novel deep learning-based approach to automatically segment the stent from IMR images, demonstrated its clinical feasibility, and validated its accuracy compared to manual segmentation. The proposed technique could be a useful tool for quantitative intrafraction motion monitoring in pancreatic cancer radiotherapy.

Evaluation of a Dedicated Software "Elements™ Spine SRS, Brainlab®" for Target Volume Definition in the Treatment of Spinal Bone Metastases With Stereotactic Body Radiotherapy.

IntroductionStereotactic body radiotherapy (SBRT) is a treatment option for spine metastases. The International Spine Radiosurgery Consortium (ISRC) has published consensus guidelines for target delineation in spine SBRT. A new software called Elements™ Spine SRS by Brainlab® that includes the module Elements SmartBrush Spine (v3.0, Munich, Germany) has been developed specifically for SBRT treatment of spine metastases, and the latter provides the ability to perform semiautomatic clinical target volume (CTV) generation based on gross tumor volume (GTV) localization and guidelines. The aims of our study were to evaluate this software by studying differences in volumes between semiautomatic CTV contours compared to manual contouring performed by an expert radiation oncologist and to determine the dosimetric impact of these differences on treatment plans.MethodsA total of 35 volumes (“Expert GTV” and “Expert CTV”) from 30 patients were defined by a single expert. A semiautomatic definition of these 35 CTVs based on the location of “Expert GTV” and following ISRC guidelines was also performed in Elements SmartBrush Spine (“Brainlab CTV”). The spatial overlap between “Brainlab” and “Expert” CTVs was calculated using the Dice similarity coefficient (DSC). We considered a threshold of 0.80 or above to indicate that Elements SmartBrush Spine performed very well with adequate contours for clinical use. Two dosimetric treatment plans, each corresponding to a specific planning target volume (PTV; Expert PTV, Brainlab PTV), were created for 11 patients.ResultsWe showed that “Brainlab CTV” and “Expert CTV” mean volumes were 29.8 ± 16.1 and 28.7 ± 15.7 cm3, respectively (p = 0.23). We also showed that the mean DSC for semiautomatic contouring relative to expert manual contouring was 0.85 ± 0.08 and less than 0.80 in five cases. For metastases involving the vertebral body only (n = 13,37%), the mean DSC was 0.90 ± 0.03, and for ones involving other or several vertebral regions (n = 22.63%), the mean DSC was 0.81 ± 0.08 (p < 0.001). The comparison of dosimetric treatment plans was performed for equivalent PTV coverage. There were no differences between doses received by organs at risk (spinal cord and esophagus) for Expert and Brainlab PTVs, respectively.ConclusionThe results showed that the semiautomatic method had quite good accuracy and can be used in clinical routine even for complex lesions.

The Learning Curve and Inter-Observer Variability in Contouring the Hippocampus under the Hippocampal Sparing Guidelines of Radiation Therapy Oncology Group 0933.

Hippocampal-sparing brain radiotherapy (HS-BRT) in cancer patients results in preservation of neurocognitive function after brain RT which can contribute to patients’ quality of life (QoL). The crucial element in HS-BRT treatment planning is appropriate contouring of the hippocampus. Ten doctors delineated the left and right hippocampus (LH and RH, respectively) on 10 patients’ virtual axial images of brain CT fused with T1-enhanced MRI (1 mm) according to the RTOG 0933 atlas recommendations. Variations in the spatial localization of the structure were described in three directions: right–left (X), cranio-caudal (Y), and forward–backward (Z). Discrepancies concerned three-dimensional localization, shape, volume and size of the hippocampus. The largest differences were observed in the first three delineated cases which were characterized by larger hippocampal volumes than the remaining seven cases. The volumes of LH of more than half of hippocampus contours were marginally bigger than those of RH. Most differences in delineation of the hippocampus were observed in the area of the posterior horn of the lateral ventricle. Conversely, a large number of hippocampal contours overlapped near the brainstem and the anterior horn of the lateral ventricle. The most problematic area of hippocampal contouring is the posterior horn of the lateral ventricle. Training in the manual contouring of the hippocampus during HS-BRT treatment planning under the supervision of experienced radiation oncologists is necessary to achieve optimal outcomes. This would result in superior outcomes of HS-BRT treatment and improvement in QoL of patients compared to without HS-BRT procedure. Correct delineation of the hippocampus is problematic. This study demonstrates difficulties in HS-BRT treatment planning and highlights critical points during hippocampus delineation.

Medical lesion segmentation by combining multimodal images with modality weighted UNet.

Automatic segmentation of medical lesions is a prerequisite for efficient clinic analysis. Segmentation algorithms for multimodal medical images have received much attention in recent years. Different strategies for multimodal combination (or fusion), such as probability theory, fuzzy models, belief functions, and deep neural networks, have also been developed. In this paper, we propose the modality weighted UNet (MW-UNet) and attention-based fusion method to combine multimodal images for medical lesion segmentation. MW-UNet is a multimodal fusion method which is based on UNet, but we use a shallower layer and fewer feature map channels to reduce the amount of network parameters, and our method uses the new multimodal fusion method called fusion attention. It uses weighted sum rule and fusion attention to combine feature maps in intermediate layers. During training, all the weight parameters are updated through backpropagation like other parameters in the network. We also incorporate residual blocks into MW-UNet to further improve segmentation performance. The comparison between the automatic multimodal lesion segmentations and the manual contours was quantified by (1) five metrics including Dice, 95% Hausdorff Distance (HD95), volumetric overlap error (VOE), relative volume difference (RVD), and mean-Intersection-over-Union (mIoU); (2) Number of parameters and flops to calculate the complexity of thenetwork. The proposed method is verified on ZJCHD, which is the data set of contrast-enhanced computed tomography (CECT) for Liver Lesion Segmentation taken from Cancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital), Hangzhou, China. For accuracy evaluation, we use 120 patients with liver lesions from ZJCHD, of which 100 are used for fourfold cross-validation (CV) and 20 are used for hold-out (HO) test. The mean Dice was and for HO and CV tests, respectively. The corresponding HD95, VOE, RVD, and mIoU of the two tests are 1.95 ± 1.83 and 2.67 ± 3.35 mm, 13.11 ± 15.83 and , 12.20 ± 18.20 and , and 83.79 ± 15.83 and . The parameters and flops of our method is 4.04 M and 18.36 G,respectively. The results show that our method performs well on multimodal liver lesion segmentation. It can be easily extended to other multimodal data sets and other networks for multimodal fusion. Our method is the potential to provide doctors with multimodal annotations and assist them with clinicaldiagnosis.

Brain tumor segmentation based on region of interest-aided localization and segmentation U-Net.

Since magnetic resonance imaging (MRI) has superior soft tissue contrast, contouring (brain) tumor accurately by MRI images is essential in medical image processing. Segmenting tumor accurately is immensely challenging, since tumor and normal tissues are often inextricably intertwined in the brain. It is also extremely time consuming manually. Late deep learning techniques start to show reasonable success in brain tumor segmentation automatically. The purpose of this study is to develop a new region-of-interest-aided (ROI-aided) deep learning technique for automatic brain tumor MRI segmentation. The method consists of two major steps. Step one is to use a 2D network with U-Net architecture to localize the tumor ROI, which is to reduce the impact of normal tissue’s disturbance. Then a 3D U-Net is performed in step 2 for tumor segmentation within identified ROI. The proposed method is validated on MICCAI BraTS 2015 Challenge with 220 high Gliomas grade (HGG) and 54 low Gliomas grade (LGG) patients’ data. The Dice similarity coefficient and the Hausdorff distance between the manual tumor contour and that segmented by the proposed method are 0.876 ±0.068 and 3.594±1.347 mm, respectively. These numbers are indications that our proposed method is an effective ROI-aided deep learning strategy for brain MRI tumor segmentation, and a valid and useful tool in medical image processing.

A novel solution of deep learning for endoscopic ultrasound image segmentation: enhanced computer aided diagnosis of gastrointestinal stromal tumor

Gastrointestinal stromal tumor is one of the critical tumors that doctors do not suggest to get frequent endoscopy, so there is a need for a diagnosis system which can process ultrasound images and figure out the tumor. Many gastrointestinal tumor diagnosis methods were developed, but all of these methods used manual contour rather than automatic segmentation. The research adopts enhanced automatic segmentation to improve the diagnosis of the gastrointestinal stromal tumor with deep convolutional neural networks. This solution’s proposed system is an enhanced automated segmentation methodology using multi-scale Gaussian kernel fuzzy clustering and multi-scale vector field convolution, which segments the ultrasound image automatically into the region of interest (the infected area). Convolutional Neural Network with Class Activation Mapping is done to diagnose an image with the tumor for Four datasets, namely (USS1, SH Hospital, SNUH, BUSI). This proposed system helps to get a clearer tumor image, and the accuracy has increased from 84.275% to 88.4%, and the processing time has reduced from 28.525% to 24.575%. The proposed solution enhanced Automatic Segmentation helped to get clearer tumor image which resulted in increased accuracy and decreased performance time compared to the state-of-the-art. Automatic segmentation overcomes the dependency on the expert for drawing the Region of Interest (ROI).

An Improved 3D Deep Learning-Based Segmentation of Left Ventricular Myocardial Diseases from Delayed-Enhancement MRI with Inclusion and Classification Prior Information U-Net (ICPIU-Net).

Accurate segmentation of the myocardial scar may supply relevant advancements in predicting and controlling deadly ventricular arrhythmias in subjects with cardiovascular disease. In this paper, we propose the architecture of inclusion and classification of prior information U-Net (ICPIU-Net) to efficiently segment the left ventricle (LV) myocardium, myocardial infarction (MI), and microvascular-obstructed (MVO) tissues from late gadolinium enhancement magnetic resonance (LGE-MR) images. Our approach was developed using two subnets cascaded to first segment the LV cavity and myocardium. Then, we used inclusion and classification constraint networks to improve the resulting segmentation of the diseased regions within the pre-segmented LV myocardium. This network incorporates the inclusion and classification information of the LGE-MRI to maintain topological constraints of pathological areas. In the testing stage, the outputs of each segmentation network obtained with specific estimated parameters from training were fused using the majority voting technique for the final label prediction of each voxel in the LGE-MR image. The proposed method was validated by comparing its results to manual drawings by experts from 50 LGE-MR images. Importantly, compared to various deep learning-based methods participating in the EMIDEC challenge, the results of our approach have a more significant agreement with manual contouring in segmenting myocardial diseases.

Autosegmentation of lung computed tomography datasets using deep learning U-Net architecture.

Current radiotherapy treatment techniques require a large amount of imaging data for treatment planning which demand significant clinician's time to segment target volume and organs at risk (OARs). In this study, we propose to use U-net-based architecture to segment OARs commonly encountered in lung cancer radiotherapy. Four U-Net OAR models were generated and trained on 20 lung cancer patients' computed tomography (CT) datasets, with each trained for 100 epochs. The model was tested for each OAR, including the right lung, left lung, heart, and spinal cord. Dice similarity coefficient (DSC) and Hausdorff distance (HD) were used to assess the agreement between the predicted contour and ground truth. The highest of the average DSC among the test patients for the left lung and the right lung was 0.96 ± 0.03 and 0.94 ± 0.06, respectively, and 0.88 ± 0.04 for heart, and 0.76 ± 0.07 for the spinal cord. The HD for these corresponding DSCs was 3.51 ± 0.85, 4.06 ± 1.12, 4.09 ± 0.85, and 2.76 ± 0.52 mm for left lung, right lung, heart, and spinal cord, respectively. The autosegmented regions predicted by right and left lung models matched well with the manual contours. However, in a few cases, the heart model struggled to outline the boundary precisely. The spinal cord model had the lowest DSC, which may be due to its small size. This is an ongoing study aimed to assist radiation oncologists in segmenting the OARs with minimal effort.

Development of Simplified Auto-Segmentable Functional Cardiac Atlas

PurposeThere is increasing evidence that radiation doses to cardiac substructures are associated with cardiac adverse events. Manual delineation of cardiac substructures is time-consuming, and auto-segmentation of cardiac substructure atlases has consequently been evaluated. However, proper automatic delineation of small substructures, such as the left anterior descending coronary artery, is challenging, and auto-segmentation of cardiac conduction system substructures has never been evaluated, despite multiple reports of radiation-induced arrhythmia after thoracic irradiations. The aim of this study was to propose and evaluate a simplified auto-segmentable functional cardiac atlas. Methods and MaterialsWe created a cardiac substructure atlas based on 20 computed tomography scans from patients with breast cancer comprising the 4 cardiac cavities, a high-risk cardiac zone as a left anterior descending coronary artery surrogate, and the 2 cardiac conduction nodes. Automatic delineation of this atlas by an atlas-based auto-segmentation algorithm was evaluated on a validation data set, consisting of 20 additional computed tomography scans. Dice similarity coefficients were used to evaluate the concordance level between the manual and the automatic contours; a dosimetric comparison between mean and maximum doses to the manual and to the auto-segmented substructures was additionally performed, based on intensity modulated radiation therapy treatment plans of the patients of the validation set. ResultsAverage dice similarity coefficient values were 0.78 for the 4 cardiac cavities, 0.65 for the high-risk cardiac zones, 0.56 for the sinoatrial node, and 0.15 for the atrioventricular node. Compared with manual contours, auto-segmented substructures were slightly smaller but the dosimetric parameters were similar. ConclusionsWe proposed a simplified functional cardiac atlas that included the cardiac conduction system and circumvented coronary delineation difficulties by using a surrogate high-risk cardiac zone. Most cardiac substructures were associated with acceptable atlas-based auto-segmentation properties. Such an atlas could be used for epidemiologic studies and for clinical practice.

Abstract P3-19-14: Development of simplified auto-segmentable functional cardiac atlas

Abstract Objective: There are increasing evidences that radiation doses to specific cardiac substructures are associated with cardiac adverse events. Manual delineation of cardiac substructures is time consuming, and auto-segmentation of cardiac substructure atlases has consequently been evaluated. However, proper delineation of small substructures, such as the left anterior descending coronary artery (LADCA), is challenging and conduction system substructures have never been considered, despite frequent reports of radiation-induced arrhythmias for thoracic irradiation. The aim of this study was to propose and evaluate a simplified auto-segmentable functional cardiac atlas. Methods: We created a cardiac substructure atlas from 20 breast cancer patients’ CT scans consisting of the four cardiac cavities, of a high-risk cardiac zone (HRCZ) as a LADCA surrogate and of the two cardiac conduction nodes. Performance evaluation of atlas-based auto-segmentation (ABAS) was evaluated on a validation data set consisting of 20 additional CT scans. Dice similarity coefficients (DSC) were used to evaluate the concordance level between the manual and the automatic segmentations. Results: The average duration of manual segmentation of the proposed cardiac atlas ranged between fifteen to twenty minutes, while ABAS lasted two minutes. The median DSC for the delineated cardiac substructures was 0.718. The highest similarity between manual and automatic segmentation was observed for the left ventricle with a median DSC of 0.87, while the lowest similarity between manual and automatic segmentation was observed for the NAV with a median DSC of 0.15. Regardless of the considered cardiac substructure, auto-segmentation tended to result into smaller volumes than manual segmentation. While the volumes were not statistically different for the left ventricle, the volumes were statistically different for the other substructures. While smaller, the auto-segmented NAV and SAN were systematically localized within the manual contours. The auto-segmented NAV could be approximated by a 1.6-cm sphere and the auto-segmented SAN by a 1.0-cm sphere. Conclusion: We proposed a simplified functional cardiac atlas, for which coronary delineation difficulties were circumvent using a surrogate high risk cardiac zone and cardiac conduction system was considered. Most cardiac substructures were associated with acceptable ABAS properties. Such atlas could be potentially evaluated for epidemiological studies and clinical practice. Table 1.performance of atlas-based auto-segmentation of the proposed simplified cardiac atlasCardiac substructureDSC Overlap index [range]Cardiac cavitiesLeft ventricle0.87 [0.57-0.92]Right ventricle0.76 [0.45-0.85]Left atria0.77 [0.61-0.84]Right atria0.76 [0.54-0.85]High risk cardiac zone0.65 [0.38-0.79]Cardiac conduction systemSinoatrial node0.56 [0.27-0.67]Atrio-ventricular node0.15 [0.00-0.35] Citation Format: Pierre Loap, Ludovic De Marzi, Krassen Kirov, Vincent Servois, Alain Fourquet, Abdelhafidh Khoubeyb, Youlia Kirova. Development of simplified auto-segmentable functional cardiac atlas [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P3-19-14.