RESUMEN
Automated colorectal cancer (CRC) segmentation in medical imaging is the key to achieving automation of CRC detection, staging, and treatment response monitoring. Compared with magnetic resonance imaging (MRI) and computed tomography colonography (CTC), conventional computed tomography (CT) has enormous potential because of its broad implementation, superiority for the hollow viscera (colon), and convenience without needing bowel preparation. However, the segmentation of CRC in conventional CT is more challenging due to the difficulties presenting with the unprepared bowel, such as distinguishing the colorectum from other structures with similar appearance and distinguishing the CRC from the contents of the colorectum. To tackle these challenges, we introduce DeepCRC-SL, the first automated segmentation algorithm for CRC and colorectum in conventional contrast-enhanced CT scans. We propose a topology-aware deep learning-based approach, which builds a novel 1-D colorectal coordinate system and encodes each voxel of the colorectum with a relative position along the coordinate system. We then induce an auxiliary regression task to predict the colorectal coordinate value of each voxel, aiming to integrate global topology into the segmentation network and thus improve the colorectum's continuity. Self-attention layers are utilized to capture global contexts for the coordinate regression task and enhance the ability to differentiate CRC and colorectum tissues. Moreover, a coordinate-driven self-learning (SL) strategy is introduced to leverage a large amount of unlabeled data to improve segmentation performance. We validate the proposed approach on a dataset including 227 labeled and 585 unlabeled CRC cases by fivefold cross-validation. Experimental results demonstrate that our method outperforms some recent related segmentation methods and achieves the segmentation accuracy in DSC for CRC of 0.669 and colorectum of 0.892, reaching to the performance (at 0.639 and 0.890, respectively) of a medical resident with two years of specialized CRC imaging fellowship.
RESUMEN
In order to understand the development law of water-conducting fractures in overlying strata during the mining process of coal seam, an elastic wave exploration method based on key stratum theory is proposed to predict the height of water-conducting fracture zone. Taking Yushen mining area as the background, the development and evolution of fractures and the three-dimensional distribution characteristics of water-conducting fracture zone are studied by combining well-ground microseismic monitoring, high-density three-dimensional seismic exploration, borehole investigation, FLAC3D numerical simulation and similar physical simulation tests. The results indicate that the trial mining face's fracture-to-coal ratio ranges from 25.86 to 30.76, with the maximum fracture-to-coal ratio near the cutting eye at 30.76 and the minimum in the central portion of the trial mining face at 25.86. The primary characteristics of rock mass fracture distribution in the mined area are the development of fractures predominantly along high-angle and even vertical bedding planes. Within the fracture zone, fractures increase from top to bottom, with high-angle fractures developing in the lower section and high-angle and horizontal fractures developing simultaneously in the upper section. The water-conducting fracture zone undergoes a developmental process from inception to development, reaching its maximum height, and eventually stabilizing as coal seam mining progresses, overlying rock subsides, strata separation, and damage formation. The three-dimensional shape of the water-conducting fracture zone in the roof of the Yushen mining area exhibits a morphological pattern where the height of the fracture zone gradually decreases from the cutting eye towards the goaf. It also transitions from high to low along both sides and from the periphery towards the interior of the working face. In the trend and strike directions, it exhibits saddle-like characteristics. By comparing the monitoring results, the rationality of the elastic wave prospecting method for predicting the height of water-conducting fracture zones based on critical layer theory was verified. This research holds significant reference value for coal mining under similar geological conditions, especially in terms of water preservation during mining operations.
RESUMEN
Brain tumor segmentation (BTS) in magnetic resonance image (MRI) is crucial for brain tumor diagnosis, cancer management and research purposes. With the great success of the ten-year BraTS challenges as well as the advances of CNN and Transformer algorithms, a lot of outstanding BTS models have been proposed to tackle the difficulties of BTS in different technical aspects. However, existing studies hardly consider how to fuse the multi-modality images in a reasonable manner. In this paper, we leverage the clinical knowledge of how radiologists diagnose brain tumors from multiple MRI modalities and propose a clinical knowledge-driven brain tumor segmentation model, called CKD-TransBTS. Instead of directly concatenating all the modalities, we re-organize the input modalities by separating them into two groups according to the imaging principle of MRI. A dual-branch hybrid encoder with the proposed modality-correlated cross-attention block (MCCA) is designed to extract the multi-modality image features. The proposed model inherits the strengths from both Transformer and CNN with the local feature representation ability for precise lesion boundaries and long-range feature extraction for 3D volumetric images. To bridge the gap between Transformer and CNN features, we propose a Trans&CNN Feature Calibration block (TCFC) in the decoder. We compare the proposed model with six CNN-based models and six transformer-based models on the BraTS 2021 challenge dataset. Extensive experiments demonstrate that the proposed model achieves state-of-the-art brain tumor segmentation performance compared with all the competitors.
Asunto(s)
Neoplasias Encefálicas , Insuficiencia Renal Crónica , Humanos , Neoplasias Encefálicas/diagnóstico por imagen , Encéfalo , Algoritmos , Calibración , Procesamiento de Imagen Asistido por ComputadorRESUMEN
Ultrasonography is an important routine examination for breast cancer diagnosis, due to its non-invasive, radiation-free and low-cost properties. However, the diagnostic accuracy of breast cancer is still limited due to its inherent limitations. Then, a precise diagnose using breast ultrasound (BUS) image would be significant useful. Many learning-based computer-aided diagnostic methods have been proposed to achieve breast cancer diagnosis/lesion classification. However, most of them require a pre-define region of interest (ROI) and then classify the lesion inside the ROI. Conventional classification backbones, such as VGG16 and ResNet50, can achieve promising classification results with no ROI requirement. But these models lack interpretability, thus restricting their use in clinical practice. In this study, we propose a novel ROI-free model for breast cancer diagnosis in ultrasound images with interpretable feature representations. We leverage the anatomical prior knowledge that malignant and benign tumors have different spatial relationships between different tissue layers, and propose a HoVer-Transformer to formulate this prior knowledge. The proposed HoVer-Trans block extracts the inter- and intra-layer spatial information horizontally and vertically. We conduct and release an open dataset GDPH&SYSUCC for breast cancer diagnosis in BUS. The proposed model is evaluated in three datasets by comparing with four CNN-based models and three vision transformer models via five-fold cross validation. It achieves state-of-the-art classification performance (GDPH&SYSUCC AUC: 0.924, ACC: 0.893, Spec: 0.836, Sens: 0.926) with the best model interpretability. In the meanwhile, our proposed model outperforms two senior sonographers on the breast cancer diagnosis when only one BUS image is given (GDPH&SYSUCC-AUC ours: 0.924 vs. reader1: 0.825 vs. reader2: 0.820).
Asunto(s)
Neoplasias de la Mama , Femenino , Humanos , Neoplasias de la Mama/diagnóstico por imagen , Ultrasonografía , Ultrasonografía Mamaria , Diagnóstico por Computador/métodosRESUMEN
This study proposes a novel idea of the use of coal gangue (CG) activation and preheated decarburized activated coal CG-based cemented paste backfill material (PCCPB) to realize green mining. PCCPB was prepared with preheated decarburized coal CG (PCG), FA, activator, low-dose cement, and water. This idea realized scale disposal and resource utilization of coal CG solid waste. Decarbonization and activation of CG crushed the material to less than 8 mm by preheated combustion technology at a combustion temperature of 900 °C and a decarbonization activation time of 4 min. The mechanism of the effect of different Na2SO4 dosages on the performance of PCCPB was investigated using comprehensive tests (including mechanical property tests, microscopic tests, and leaching toxicity tests). The results show that the uniaxial compressive strength (UCS) of C-S2, C-S3, and C-S4 can meet the requirements of backfill mining, among which the UCS of C-S3 with a curing time of 3 d and 28 d were 0.545 MPa and 4.312 MPa, respectively. Na2SO4 excites PCCPB at different curing time, and the UCS of PCCPB increases and then decreases with the increase in Na2SO4 dosage, and 3% of Na2SO4 had the best excitation effect on the late strength (28 d) of PCCPB. All groups' (control and CS1-CS4 groups) leachate heavy metal ions met the requirements of groundwater class III standard, and PCCPB had a positive effect on the stabilization/coagulation of heavy metal ions (Mn, Zn, As, Cd, Hg, Pb, Cr, Ba, Se, Mo, and Co). Finally, the microstructure of PCCPB was analyzed using FTIR, TG/DTG, XRD, and SEM. The research is of great significance to promote the resource utilization of coal CG residual carbon and realize the sustainable consumption of coal CG activation on a large scale.
RESUMEN
With the wide application of the filling mining method, it is necessary to consider the influence of rock activity on the filling body, reflected in the laboratory, that is, the influence of loading rate. Therefore, to explore the response characteristics of loading rate on the mechanical and damage characteristics of aeolian sand paste filling body, DNS100 electronic universal testing machine and DS5-16B acoustic emission (AE) monitoring system were used to monitor the stress-strain changes and AE characteristic parameters changes of aeolian sand paste filling body during uniaxial compression, and the theoretical model of filling sample damage considering loading rate was established based on AE parameters. The experimental results show that: (1) With the increase in loading rate, the uniaxial compressive strength and elastic modulus of aeolian sand paste-like materials (ASPM) specimens are significantly improved. ASPM specimens have ductile failure characteristics, and the failure mode is unidirectional shear failure â tensile failure â bidirectional shear failure. (2) When the loading rate is low, the AE event points of ASPM specimens are more dispersed, and the large energy points are less. At high loading rates, the AE large energy events are more concentrated in the upper part, and the lower part is more distributed. (3) The proportion of the initial active stage is negatively correlated with the loading rate, and the proportion of the active stage is positively correlated with the loading rate. The total number of AE cumulative ringing decreases with the increase in loading rate. (4) Taking time as an intermediate variable, the coupling relationship between ASPM strain considering loading rate and the AE cumulative ringing count is constructed, and the damage and stress coupling model of ASPM specimen considering loading rate is further deduced. Comparing the theoretical model with the experimental results shows that the model can effectively reflect the damage evolution process of ASPM specimens during loading, especially at high loading rates. The research results have significant reference value for subsequent strength design of filling material, selection of laboratory loading rate and quality monitoring, and early warning of filling body in goaf.
RESUMEN
Cells/nuclei deliver massive information of microenvironment. An automatic nuclei segmentation approach can reduce pathologists' workload and allow precise of the microenvironment for biological and clinical researches. Existing deep learning models have achieved outstanding performance under the supervision of a large amount of labeled data. However, when data from the unseen domain comes, we still have to prepare a certain degree of manual annotations for training for each domain. Unfortunately, obtaining histopathological annotations is extremely difficult. It is high expertise-dependent and time-consuming. In this paper, we attempt to build a generalized nuclei segmentation model with less data dependency and more generalizability. To this end, we propose a meta multi-task learning (Meta-MTL) model for nuclei segmentation which requires fewer training samples. A model-agnostic meta-learning is applied as the outer optimization algorithm for the segmentation model. We introduce a contour-aware multi-task learning model as the inner model. A feature fusion and interaction block (FFIB) is proposed to allow feature communication across both tasks. Extensive experiments prove that our proposed Meta-MTL model can improve the model generalization and obtain a comparable performance with state-of-the-art models with fewer training samples. Our model can also perform fast adaptation on the unseen domain with only a few manual annotations. Code is available at https://github.com/ChuHan89/Meta-MTL4NucleiSegmentation.
Asunto(s)
Procesamiento de Imagen Asistido por Computador , Redes Neurales de la Computación , Algoritmos , HumanosRESUMEN
Tissue-level semantic segmentation is a vital step in computational pathology. Fully-supervised models have already achieved outstanding performance with dense pixel-level annotations. However, drawing such labels on the giga-pixel whole slide images is extremely expensive and time-consuming. In this paper, we use only patch-level classification labels to achieve tissue semantic segmentation on histopathology images, finally reducing the annotation efforts. We propose a two-step model including a classification and a segmentation phases. In the classification phase, we propose a CAM-based model to generate pseudo masks by patch-level labels. In the segmentation phase, we achieve tissue semantic segmentation by our propose Multi-Layer Pseudo-Supervision. Several technical novelties have been proposed to reduce the information gap between pixel-level and patch-level annotations. As a part of this paper, we introduce a new weakly-supervised semantic segmentation (WSSS) dataset for lung adenocarcinoma (LUAD-HistoSeg). We conduct several experiments to evaluate our proposed model on two datasets. Our proposed model outperforms five state-of-the-art WSSS approaches. Note that we can achieve comparable quantitative and qualitative results with the fully-supervised model, with only around a 2% gap for MIoU and FwIoU. By comparing with manual labeling on a randomly sampled 100 patches dataset, patch-level labeling can greatly reduce the annotation time from hours to minutes. The source code and the released datasets are available at: https://github.com/ChuHan89/WSSS-Tissue.
Asunto(s)
Procesamiento de Imagen Asistido por Computador , Aprendizaje Automático Supervisado , Humanos , Procesamiento de Imagen Asistido por Computador/métodos , SemánticaRESUMEN
Nuclei segmentation is a vital step for pathological cancer research. It is still an open problem due to some difficulties, such as color inconsistency introduced by non-uniform manual operations, blurry tumor nucleus boundaries and overlapping tumor cells. In this paper, we aim to leverage the unique optical characteristic of H&E staining images that hematoxylin always stains cell nuclei blue, and eosin always stains the extracellular matrix and cytoplasm pink. Therefore, we extract the Hematoxylin component from RGB images by Beer-Lambert's Law. According to the optical attribute, the extracted Hematoxylin component is robust to color inconsistency. With the Hematoxylin component, we propose a Hematoxylin-aware CNN model for nuclei segmentation without the necessity of color normalization. Our proposed network is formulated as a Triple U-net structure which includes an RGB branch, a Hematoxylin branch and a Segmentation branch. Then we propose a novel feature aggregation strategy to allow the network to fuse features progressively and to learn better feature representations from different branches. Extensive experiments are performed to qualitatively and quantitatively evaluate the effectiveness of our proposed method. In the meanwhile, it outperforms state-of-the-art methods on three different nuclei segmentation datasets.
Asunto(s)
Núcleo Celular , Redes Neurales de la Computación , Eosina Amarillenta-(YS) , Hematoxilina , Coloración y EtiquetadoRESUMEN
An enteroatmospheric fistula is a devastating complication in the open abdomen. Usually the fistula forms before the completion of split-thickness skin graft surgery. A 35-year-old woman admitted with pancreatic and liver injuries, and postoperative sepsis underwent open abdomen treatment and developed two enteroatmospheric fistulae 14 days after split-thickness skin grafting. The complication was believed to have occurred as a result of multiple surgical manipulations for intra-abdominal hemorrhage and skin graft dressing changes. One fistula, measuring 0.5 cm in diameter, was managed using a tailored 20-mL syringe, secured to the surrounding tissues with ostomy paste, and a suction catheter. The other fistula, measuring 1.8 cm in diameter, required insertion of a catheter to collect the effluent. Once the effluent was controlled effectively, a second split-thickness skin graft procedure was performed to facilitate fistula management. The patient remained stable until successful fistula repair 8 months later, and she is now awaiting elective abdominal wall reconstruction. This case study is an important reminder that patients with an open abdomen, even after split-thickness skin grafting, are at risk for enteroatmospheric fistula formation. Once this severe complication occurs, effective control of fistula effluent and subsequent split-thickness skin grafting procedures are needed.