Lobe Segment Research Articles

DDA-SSNets: Dual decoder attention-based semantic segmentation networks for COVID-19 infection segmentation and classification using chest X-Ray images.

COVID-19 needs to be diagnosed and staged to be treated accurately. However, prior studies' diagnostic and staging abilities for COVID-19 infection needed to be improved. Therefore, new deep learning-based approaches are required to aid radiologists in detecting and quantifying COVID-19-related lung infections. To develop deep learning-based models to classify and quantify COVID-19-related lung infections. Initially, Dual Decoder Attention-based Semantic Segmentation Networks (DDA-SSNets) such as Dual Decoder Attention-UNet (DDA-UNet) and Dual Decoder Attention-SegNet (DDA-SegNet) are proposed to facilitate the dual segmentation tasks such as lung lobes and infection segmentation in chest X-ray (CXR) images. The lung lobe and infection segmentations are mapped to grade the severity of COVID-19 infection in both the lungs of CXRs. Later, a Genetic algorithm-based Deep Convolutional Neural Network classifier with the optimum number of layers, namely GADCNet, is proposed to classify the extracted regions of interest (ROI) from the CXR lung lobes into COVID-19 and non-COVID-19. The DDA-SegNet shows better segmentation with an average BCSSDC of 99.53% and 99.97% for lung lobes and infection segmentations, respectively, compared with DDA-UNet with an average BCSSDC of 99.14% and 99.92%. The proposed DDA-SegNet with GADCNet classifier offered excellent classification results with an average BCCAC of 99.98%, followed by the GADCNet with DDA-UNet with an average BCCAC of 99.92% after extensive testing and analysis. The results show that the proposed DDA-SegNet has superior performance in the segmentation of lung lobes and COVID-19-infected regions in CXRs, along with improved severity grading compared to the DDA-UNet and improved accuracy of the GADCNet classifier in classifying the CXRs into COVID-19, and non-COVID-19.

Metastatic pathways to the lower zone by segment in patients with clinical T1 lower lobe non-small cell lung cancer.

Segmentectomy and mediastinal lymph node dissection (LND) may increasingly be used for non-small cell lung cancer (NSCLC). Lymph node metastasis (LNM) distribution varies by lower lobe segments; however, its segment-specific spread to the lower zone (#8, 9) (LZ) in lower lobe NSCLC is seldom reported. In total, 352 patients with clinical T1 lower lobe NSCLC who underwent lobectomy with systematic or lobe-specific LND were included for analysis between January 2006 and December 2018. Fifty-eight (16.2%) patients had LNM (pN1: 24, pN2: 34), and nine (2.6%) had LZ metastasis. LZ metastasis was significantly more frequent in tumors with diameter > 2cm, tumors without ground-glass opacity on radiological findings, left lung cancer, and basal segment lung cancer (respectively, p = 0.039, 0.006, 0.0177, 0.0024). None of the S6 NSCLC patients had LZ metastasis. Two patients with right basal segment NSCLC had LZ metastases (tumor on S10) as well as N1 lymph node and subcarinal zone metastasis. Seven (8.4%) patients with left basal segment NSCLC had LZ metastasis (tumor on S8: 3, tumor on S10: 4). Of them, three patients with left basal NSCLC had isolated LZ metastasis. The LND of the LZ can be omitted for clinical T1 patients with S6 NSCLC. In addition, the LND of the LZ may be omitted in right basal NSCLC if intraoperative confirmation of negative N1 and subcarinal zone lymph nodes is obtained; however, it is necessary for left basal segment NSCLC.

Solitary Pulmonary Nodule Under the Anonymity of COVID-19 Infection

Abstract We report the case of a 32-year-old male patient, a former smoker, with a known history of SARS-COV-2 infection, mild clinical form, condition that required follow-up evaluation. Despite being oligosymptomatic, the patient reports discreet respiratory symptoms, manifested by dry cough and a sensation of chest tightness. CT thoracic scan revealed a pseudo-nodular, diffusely contoured infiltrate of approximately 10 mm, in the apical left upper lobe segment (LUL), sequelae appearance, thought to represent SARS-COV-2 residual changes, with follow-up recommendation. Three months later, reassessment highlighted a slight progression of the lesion, with a spiculated appearance and umbilicated pleura, a context that mandated further respiratory-specific medical investigations, including bronchoscopic examination and positron emission tomography (PET CT). Given the diagnostic suspicion that included a pathology with neoplastic potential, the firm recommendation was to perform a nodular pulmonary biopsy through video-assisted thoracoscopy (VATS). Both the histopathological (HP) and immunohistochemical (IHC) examinations supported the diagnosis of primary pulmonary adenocarcinoma. The therapeutic course was completed by performing a left upper lobectomy with lymphadenectomy. The case is instructive, confirming that SARS-COV-2 infection is a perfect mimic (1), capable of imitating various other conditions. In this context, the optimal scenario is distinguished, where monitoring after infection contributes to the early and incidental identification of neoplasia, thus suggesting that SARS-COV-2 infection may also offer certain benefits.

Pulmonary Artery Wedge Pressure Can Underestimate Direct Pulmonary Vein Pressure in Pediatric Pulmonary Vein Stenosis

BackgroundPulmonary artery wedge pressure (PAWP) can underestimate directly measured pulmonary vein pressure (PVP) as demonstrated in animal studies and human case reports. This concept has not been validated in a larger cohort of pediatric patients with pulmonary vein stenosis (PVS). MethodsPediatric patients who underwent cardiac catheterization for PVS at a single center from January 1, 2018, to March 31, 2023, were retrospectively reviewed. Mismatch between the PAWP and directly measured PVP or LA pressure was defined as >3 mm Hg. Preintervention angiography was reviewed and percent stenosis calculated. ResultsTwenty-six patients met inclusion criteria; 51 lower pulmonary veins (34 left, 17 right) from 42 catheterizations were evaluated. Significant PVS (≥30% stenosis) was seen in 38/51 (75%) veins, and 9/51 (18%) veins had no angiographic narrowing (0% stenosis). PAWP-PVP mismatch occurred in 37/51 (73%) veins with a median difference of 8 mm Hg (IQR, 6-12). Of these, PAWP was equal to LA pressure in 26 instances, all of which had significant PVS (median % stenosis 54 [IQR, 45-60]). Six of the cases with PAWP-PVP mismatch, PVS (range, 41%-70% stenosis), and no PAWP-LA mismatch reported both a proximal and distal segmental PAWP. In all 6 instances, there was no distal PAWP-PVP mismatch (median difference 1 mm Hg [range, 0-3]). ConclusionsIn this single-center study of pediatric patients with PVS, PAWP significantly underestimated directly measured PVP in lower pulmonary veins. Balloon wedge or end hole catheter position in a distal lobar segment may more accurately estimate the PVP.

Uniportal VATS left lower lobe and lingual segment sleeve resection

Uniportal VATS left lower lobe and lingual segment sleeve resection

Lightweight Techniques to Improve Generalization and Robustness of U-Net Based Networks for Pulmonary Lobe Segmentation.

Lung lobe segmentation in chest CT is relevant to a wide range of clinical applications. However, existing segmentation pipelines often exhibit vulnerabilities and performance degradations when applied to external datasets. This is usually attributed to the size of the available dataset or model. We show that it is possible to enhance generalizability without huge resources by carefully curating the dataset and combining machine learning with medical expertise. Multiple machine learning techniques (self-supervision (SSL), attention (A), and data augmentation (DA)) are used to train a fast and fully-automated lung lobe segmentation model based on 2D U-Net. Our study involved evaluating these techniques on a diverse dataset collected under the RACOON project, encompassing 100 CT chest scans from patients with bacterial, viral, or SARS-CoV2 infections. We compare our model to a baseline U-Net trained on the same dataset. Our approach significantly improved segmentation accuracy (Dice score of 92.8% vs. 82.3%, p < 0.001). Moreover, our model achieved state-of-the-art performance (Dice score of 92.8% vs. 90.8% for the literature's state-of-the-art, p = 0.102) with reduced training examples (69 vs. 231 CT Scans). Among the techniques, data augmentation with expert knowledge displayed the most significant impact, enhancing the Dice score by +0.056. Notably, these enhancements are not limited to lobe segmentation but can be seamlessly integrated into various medical imaging segmentation tasks, demonstrating their versatility and potential for broader applications.

A Novel Air Trapping Segment Score Identifies Opposing Effects of Obesity and Eosinophilia on Air Trapping in Asthma.

Density thresholds in computed tomography (CT) lung scans quantify air trapping (AT) at the whole lung level but are not informative for AT in specific bronchopulmonary segments. To apply a segment-based measure of AT in asthma to investigate the clinical determinants of AT in asthma. In each of 19 bronchopulmonary segments in CT lung scans from 199 asthma patients, AT was categorized as present if lung attenuation was < -856 Hounsfield units at expiration in ≥ 15% of lung area. The resulting AT segment score (0-19) was related to patient outcomes. AT varied at the lung segment level and tended to persist at the patient and lung segment level over 3 years. Patients with widespread AT (>10 segments) had more severe asthma (p<0.05). The mean (± SD) AT segment score in patients with a BMI > 30 was lower than in patients with a BMI < 30 (3.5 ± 4.6 vs. 5.5 ± 6.3, p=0.008), and the frequency of AT in lower lobe segments in obese patients was less than in upper and middle lobe segments (35 vs. 46%, p=0.001). The AT segment score in patients with sputum eosinophil % > 2 was higher than in patients without sputum eosinophilia (7.0 ± 6.1 vs. 3.3 ± 4.9, p<0.0001). Lung segments with AT more frequently had airway mucus plugging than lung segments without AT (48 vs. 18%, p≤0.0001). In patients with asthma, air trapping is more severe in those with airway eosinophilia and mucus plugging whereas those who are obese have less severe trapping because their lower lobe segments are spared.

NmPLS-Net: Segmenting Pulmonary Lobes Using nmODE

Pulmonary lobe segmentation is vital for clinical diagnosis and treatment. Deep neural network-based pulmonary lobe segmentation methods have seen rapid development. However, there are challenges that remain, e.g., pulmonary fissures are always not clear or incomplete, especially in the complex situation of the trilobed right pulmonary, which leads to relatively poor results. To address this issue, this study proposes a novel method, called nmPLS-Net, to segment pulmonary lobes effectively using nmODE. Benefiting from its nonlinear and memory capacity, we construct an encoding network based on nmODE to extract features of the entire lung and dependencies between features. Then, we build a decoding network based on edge segmentation, which segments pulmonary lobes and focuses on effectively detecting pulmonary fissures. The experimental results on two datasets demonstrate that the proposed method achieves accurate pulmonary lobe segmentation.

Hybrid optimized MRF based lung lobe segmentation and lung cancer classification using Shufflenet

Hybrid optimized MRF based lung lobe segmentation and lung cancer classification using Shufflenet

Application of FGD-BCEL loss function in segmenting temporal lobes on localized CT images for radiotherapy.

The aim of this study was to find a new loss function to automatically segment temporal lobes on localized CT images for radiotherapy with more accuracy and a solution to dealing with the classification of class-imbalanced samples in temporal lobe segmentation. Localized CT images for radiotherapy of 70 patients with nasopharyngeal carcinoma were selected. Radiation oncologists sketched mask maps. The dataset was randomly divided into the training set (n = 49), the validation set (n = 7), and the test set (n = 14). The training set was expanded by rotation, flipping, zooming, and shearing, and the models were evaluated using Dice similarity coefficient (DSC), Jaccard similarity coefficient (JSC), positive predictive value (PPV), sensitivity (SE), and Hausdorff distance (HD). This study presented an improved loss function, focal generalized Dice-binary cross-entropy loss (FGD-BCEL), and compared it with four other loss functions, Dice loss (DL), generalized Dice loss (GDL), Tversky loss (TL), and focal Tversky loss (FTL), using the U-Net model framework. With the U-Net model based on FGD-BCEL, the DSC, JSC, PPV, SE, and HD were 0.87 ± 0.11, 0.78 ± 0.11, 0.90 ± 0.10, 0.87 ± 0.13, and 4.11 ± 0.75, respectively. Except for the SE, all the other evaluation metric values of the temporal lobes segmented by the FGD-BCEL-based U-Net model were improved compared to the DL, GDL, TL, and FTL loss function-based U-Net models. Moreover, the FGD-BCEL-based U-Net model was morphologically more similar to the mask maps. The over- and under-segmentation was lessened, and it effectively segmented the tiny structures in the upper and lower poles of the temporal lobe with a limited number of samples. For the segmentation of the temporal lobe on localized CT images for radiotherapy, the U-Net model based on the FGD-BCEL can meet the basic clinical requirements and effectively reduce the over- and under-segmentation compared with the U-Net models based on the other four loss functions. However, there still exists some over- and under-segmentation in the results, and further improvement is needed.

A radiomics-based logistic regression model for the assessment of emphysema severity.

The aim of this study is to develop a model that differentiates between the radiological patterns of severe and mild emphysema using radiomics parameters, as well as to examine the parameters included in the model. Over the last 12 months, a total of 354 patients were screened based on the presence of terms such as “Fleischner”, “CLE”, and “centriacinar” in their thoracic CT reports, culminating in a study population of 82 patients. The study population was divided into Group 1 (Fleischner mild and moderate; n= 45) and Group 2 (Fleischner confluent and advanced destructive; n= 37). Volumetric segmentation was performed, focusing on the upper lobe segments of both lungs. From these segmented volumes, radiomics parameters including shape, size, first-order, and second-order features were calculated. The best model parameters were selected based on the Bayesian Information Criterion and further optimized through grid search. The final model was tested using 1000 iterations of bootstrap resampling. In the training set, performance metrics were calculated with a sensitivity of 0.862, specificity of 0.870, accuracy of 0.863, and AUC of 0.910. Correspondingly, in the test set, these values were sensitivity= 0.848; specificity= 0.865; accuracy= 0.857; and AUC= 0.907. The logistic regression model, composed of radiomics parameters and trained on a limited number of cases, effectively differentiated between mild and severe radiological patterns of emphysema using computed tomography images.

War Snake Optimisation Algorithm with deep Q-Net for COVID-19 classification

ABSTRACT A virus known as the Coronavirus 2019 (COVID-19) can have serious effects on a variety of body parts. Preventing this spread of the virus requires early detection and proper treatment. Many methods are followed for early detection of this disease based on symptoms and classification for proper treatment. In this paper, classification of COVID-19 using Computed Tomography (CT) scan images with hybrid optimisation-enabled deep learning is researched. Here, a hybrid optimisation algorithm called War Snake Optimisation Algorithm (WSOA) is proposed to train Deep Q-Network (Deep Q-Net). Moreover, the proposed WSOA is formed by integrating War strategy optimisation (WSO) and Snake Optimiser (SO). In this paper, pre-processing is carried out using a median filter, followed by lung lobe segmentation that is done by Generative Adversarial Network (GAN). Furthermore, image augmentation is done by translation, rotation, flipping and scaling. Then, feature extraction is followed after image augmentation, where needed features get extracted to end up with the classification of COVID-19. This classification brings out the patient’s health condition of normal and abnormal situations. Moreover, this research work is analysed with three performance metrics, such as accuracy, sensitivity and specificity with values of 93.7%, 94.8% and 89.5%.

Comparison of hyperpolarized 3He-MRI, CT based parametric response mapping, and mucus scores in asthmatics.

Purpose: The purpose of this study was to anatomically correlate ventilation defects with regions of air trapping by whole lung, lung lobe, and airway segment in the context of airway mucus plugging in asthma. Methods: A total of 34 asthmatics [13M:21F, 13 mild/moderate, median age (range) of 49.5 (36.8-53.3) years and 21 severe, 56.1 (47.1-62.6) years] and 4 healthy subjects [1M:3F, 38.5 (26.6-52.2) years] underwent HP 3He MRI and CT imaging. HP 3He MRI was assessed for ventilation defects using a semi-automated k-means clustering algorithm. Inspiratory and expiratory CTs were analyzed using parametric response mapping (PRM) to quantify markers of emphysema and functional small airways disease (fSAD). Segmental and lobar lung masks were obtained from CT and registered to HP 3He MRI in order to localize ventilation defect percent (VDP), at the lobar and segmental level, to regions of fSAD and mucus plugging. Spearman's correlation was utilized to compare biomarkers on a global and lobar level, and a multivariate analysis was conducted to predict segmental fSAD given segmental VDP (sVDP) and mucus score as variables in order to further understand the functional relationships between regional measures of obstruction. Results: On a global level, fSAD was correlated with whole lung VDP (r = 0.65, p < 0.001), mucus score (r = 0.55, p < 0.01), and moderately correlated (-0.60 r -0.56, p < 0.001) to percent predicted (%p) FEV1, FEF25-75 and FEV1/FVC, and more weakly correlated to FVC%p (-0.38 r -0.35, p < 0.001) as expected from previous work. On a regional level, lobar VDP, mucus scores, and fSAD were also moderately correlated (r from 0.45-0.66, p < 0.01). For segmental colocalization, the model of best fit was a piecewise quadratic model, which suggests that sVDP may be increasing due to local airway obstruction that does not manifest as fSAD until more extensive disease is present. sVDP was more sensitive to the presence of a mucus plugs overall, but the prediction of fSAD using multivariate regression showed an interaction in the presence of a mucus plugs when sVDP was between 4% and 10% (p < 0.001). Conclusion: This multi-modality study in asthma confirmed that areas of ventilation defects are spatially correlated with air trapping at the level of the airway segment and suggests VDP and fSAD are sensitive to specific sources of airway obstruction in asthma, including mucus plugs.

Quantifying emphysema in lung screening computed tomography with robust automated lobe segmentation.

Anatomy-based quantification of emphysema in a lung screening cohort has the potential to improve lung cancer risk stratification and risk communication. Segmenting lung lobes is an essential step in this analysis, but leading lobe segmentation algorithms have not been validated for lung screening computed tomography (CT). In this work, we develop an automated approach to lobar emphysema quantification and study its association with lung cancer incidence. We combine self-supervised training with level set regularization and finetuning with radiologist annotations on three datasets to develop a lobe segmentation algorithm that is robust for lung screening CT. Using this algorithm, we extract quantitative CT measures for a cohort () from the National Lung Screening Trial and analyze the multivariate association with lung cancer incidence. Our lobe segmentation approach achieved an external validation Dice of 0.93, significantly outperforming a leading algorithm at 0.90 (). The percentage of low attenuation volume in the right upper lobe was associated with increased lung cancer incidence (odds ratio: 1.97; 95% CI: [1.06, 3.66]) independent of risk factors and diagnosis of whole lung emphysema. Quantitative lobar emphysema improved the goodness-of-fit to lung cancer incidence (, ). We are the first to develop and validate an automated lobe segmentation algorithm that is robust to smoking-related pathology. We discover a quantitative risk factor, lending further evidence that regional emphysema is independently associated with increased lung cancer incidence. The algorithm is provided at https://github.com/MASILab/EmphysemaSeg.

Learnable interpolation and extrapolation network for fuzzy pulmonary lobe segmentation

AbstractPulmonary lobe segmentation is an important prerequisite for accurately quantifying pulmonary damage in many pulmonary diseases and planning treatment. However, due to the incomplete lobar structures and morphological changes caused by diseases, the lobe segmentation still encounters great challenges. In this study, a Learnable Interpolation and Extrapolation Network (LIE‐Net) is proposed to form complete and consecutive fissure surfaces by learning to extract information of the fissures from existing fissure points and absent points (unsegmented points belonging to fissures) to predict the z coordinate of the absent fissure points. The completed pulmonary fissures are further used for accurate pulmonary lobe segmentation. Specifically, LIE‐Net takes the coordinate information of existing fissure points (their (x, y, z) coordinates) and absent fissure points (their (x, y) coordinates) as two independent inputs, and predicts the z coordinates of absent points. The proposed LIE‐Net makes voxel‐wise predictions based on the spatial structure characteristics of the lung fissure, and is able to provide a consecutive fissure surface in space. According to the evaluation of radiologists, the lobe segmentation performance was remarkably enhanced in approximately 76% of patients in our additional dataset after the application of LIE‐Net, especially for those cases with large‐scale missing fissures.

SCLMnet: A dual-branch guided network for lung and lung lobe segmentation

Lung and lung lobe segmentation are two crucial techniques for lung imaging analysis that interact in clinical settings. Lung segmentation assists physicians in comparing different images to select the most appropriate surgical plan, while lobe segmentation provides precise anatomical information to help plan surgical procedures. However, inaccurate lung segmentation edges, mis-segmented lobe boundaries, and tiny targets pose challenges. Therefore, we propose a dual-branch guided convolutional neural network, SCLMnet, for lung and lung lobe segmentation. To completely leverage the semantic information of feature maps, the first branch adds a spatial linkage module (SLM) to focus on low-level features at different spatial levels, highlighting feature representations of lung edges and lung lobe boundaries. A channel linkage module (CLM) is added by matrix inner product to model channel relations, emphasizing the relevance and similarity of feature maps and capturing the interdependency of high-level feature channels to highlight feature representations of the entire lung lobe region. Transmodal synaptic linkage (TSL) and multi-scale fusion strategy guide the feature information of the CLM and SLM and the deep features extracted by the second branch ResUNet to jointly explore useful information in chest computer tomography (CT) images. To evaluate the performance of the state-of-the-art model, we use three publicly available datasets: LUNA16, COVID-19-CT-Seg, and VESSEL12. Compared to the existing methods, SCLMnet achieves average Dice scores of 92.17%, 97.80%, and 99.12%, respectively, demonstrating remarkable performance, which suggests that lung and lung lobe segmentation using CT images with SCLMnet can play an essential role in clinical research.

Remora Namib Beetle Optimization Enabled Deep Learning for Severity of COVID-19 Lung Infection Identification and Classification Using CT Images.

Coronavirus disease 2019 (COVID-19) has seen a crucial outburst for both females and males worldwide. Automatic lung infection detection from medical imaging modalities provides high potential for increasing the treatment for patients to tackle COVID-19 disease. COVID-19 detection from lung CT images is a rapid way of diagnosing patients. However, identifying the occurrence of infectious tissues and segmenting this from CT images implies several challenges. Therefore, efficient techniques termed as Remora Namib Beetle Optimization_ Deep Quantum Neural Network (RNBO_DQNN) and RNBO_Deep Neuro Fuzzy Network (RNBO_DNFN) are introduced for the identification as well as classification of COVID-19 lung infection. Here, the pre-processing of lung CT images is performed utilizing an adaptive Wiener filter, whereas lung lobe segmentation is performed employing the Pyramid Scene Parsing Network (PSP-Net). Afterwards, feature extraction is carried out wherein features are extracted for the classification phase. In the first level of classification, DQNN is utilized, tuned by RNBO. Furthermore, RNBO is designed by merging the Remora Optimization Algorithm (ROA) and Namib Beetle Optimization (NBO). If a classified output is COVID-19, then the second-level classification is executed using DNFN for further classification. Additionally, DNFN is also trained by employing the newly proposed RNBO. Furthermore, the devised RNBO_DNFN achieved maximum testing accuracy, with TNR and TPR obtaining values of 89.4%, 89.5% and 87.5%.

Comparison of different approaches in complete thoracoscopic segmentectomy of lung lower lobe.

Safe and oncologically acceptable segmentectomy outcomes were reported for early-stage lung cancer. The high-resolution computed tomography allowed us to find detailed structures inside the lungs, such as the pulmonary ligaments (PLs). Hence, we have presented the relatively anatomically challenging thoracoscopic segmentectomy, for the resection of the lateral basal segment, the posterior basal segment, and both segments through the PL as a PL approach. This study aimed to retrospectively examine the lung lower lobe segmentectomy, excluding the superior and basal segments (from S7 to S10), using the PL approach as an option to treat the lower lobe tumors of the lung. We then compared the efficacy of the PL approach in terms of safety with the interlobar fissure (IF) approach. The characteristics of the patients, intra- and postoperative complications, and surgical outcomes were analyzed. Of the 510 patients who underwent segmentectomy for malignant lung tumors from February 2009 to December 2020, 85 were included in this study. Among them, 41 underwent a complete lung lower lobe thoracoscopic segmentectomy, excluding S6 and basal segments (from S7 to S10), using the PL approach, and the remaining 44 used the IF approach. The median age in 41 patients in the PL group was 64.0 years (range, 22-82), and that in 44 patients in the IF group was 66.5 years (range, 44-88), with significant differences in gender between these groups. Video-assisted thoracoscopic surgery and robot-assisted thoracoscopic surgery were performed on 37 and 4 patients in the PL group and 43 and 1 patient in the IF group, respectively. Postoperative complication frequency was not significantly different between these groups. The most common complications were the air leaks that persisted for over 7 days in 1 and 5 patients in the PL and IF groups, respectively. Complete thoracoscopic segmentectomy of the lower lobe, excluding S6 and basal segments, using the PL approach is a reasonable option for lung lower lobe tumors compared with the IF approach.

Lung lobe segmentation in computed tomography images based on multi‐feature fusion and ensemble learning framework

AbstractLung lobe segmentation in computed tomography (CT) images can be regarded as essential supporting information for the diagnosis and treatment of lung diseases, yet it is a challenging uncertainty for the complex segmentation task due to the diverse structures like indistinguishable pulmonary arteries and veins, unpredictable pathological deformation and blurring pulmonary fissures. To circumvent these challenges, we present a productive deep learning network based on multi‐feature fusion and ensemble learning approach to highlight pulmonary fissure representation and suppress other structures for lung lobe segmentation in CT images. Considering that pulmonary fissures are the physical boundaries of lung lobes, a multi‐feature fusion approach is presented to integrate original images, enhanced fissure magnitude images, and enhanced fissure orientation images to highlight shape representation for semantic segmentation. In addition, a deep‐supervised ensemble learning network is employed to combine multiple inducers for lung lobe segmentation improvement. The performance of the proposed deep learning model is validated in experiments with an available dataset, it acquired a high IoU value and a low Hausdorff distance compared with manual references. Experimental results demonstrate that the proposed computational model based on multi‐feature fusion and deep‐supervised ensemble learning framework has an improved predictive performance than many state‐of‐the‐art deep neural networks in lung lobe segmentation.

Stereo-EEG localization of midline onset seizures on scalp EEG

PurposeThe objective of this study was to describe the sEEG-defined seizure onset zone (SOZ), seizure semiology, presurgical evaluations, surgical intervention and outcome in patients with midline onset noninvasive phase I monitoring. MethodsA single center sEEG database was reviewed to identify patients with seizures onset predominantly involving midline electrodes (FZ, CZ, PZ, OZ) on scalp EEG. Data abstracted included clinical factors, seizure semiology graded into lobar segmentation, imaging and electrographic findings, sEEG plan, interventions, and outcome. ResultsTwelve patients were identified (8 males, median age of sEEG 28 years) out of 100 cases of sEEG performed from January 2015-September 2019. “Frontal lobe” seizure semiology was the most common. sEEG-defined SOZ were frontal (5), diffuse (1), multifocal (1), frontal and insular (1), frontal and cingulate (1), insular (1), cingulate (1), and mesial temporal (1). CZ and/or FZ scalp EEG changes were present for all patients with SOZ involving the frontal, cingulate, and insular regions. PZ/OZ scalp involvement was present in one patient with mesial temporal SOZ. Four patients underwent a definitive resective or ablative surgery, and the remaining patients underwent a palliative intervention. Of those with follow-up information available, 8/11 had seizure reduction by ≥ 50%, including 4 with an Engel I outcome. No clinical factors were associated with outcome. ConclusionsSOZ for midline onset seizures from noninvasive phase I monitoring was most commonly in the frontal, cingulate, and insular regions. A complex cortical network between these regions may explain overlap in semiology and scalp EEG findings. While the number rendered seizure-free was limited, a significant proportion experienced a reasonably favorable outcome justifying use of sEEG to identify surgical options in these patients.