Search

Cancer is a complex disease that can affect any body part. One key feature of cancer is the rapid production of abnormal cells that grow beyond their usual borders and can invade adjoining parts of the body and spread/metastasized to other organs. The process of metastasis is the crucial cause of cancer death. Environmental factors are a significant contributor to cancer initiation [1]. Numerous studies investigate various aspects of cancer, including pathogenesis, prevention, diagnosis, and treatment methods, with the goal of improving patient quality of life and increasing survival rates. Despite significant advances in the field, cancer continues to represent a global challenge for prevention and treatment.

Anatomical resection (AR) based on anatomical sub-regions is a promising method of precise surgical resection, which has been proven to improve long-term survival by reducing local recurrence. The fine-grained segmentation of an organ's surgical anatomy (FGS-OSA), i.e., segmenting an organ into multiple anatomic regions, is critical for localizing tumors in AR surgical planning. However, automatically obtaining FGS-OSA results in computer-aided methods faces the challenges of appearance ambiguities among sub-regions (i.e., inter-sub-region appearance ambiguities) caused by similar HU distributions in different sub-regions of an organ's surgical anatomy, invisible boundaries, and similarities between anatomical landmarks and other anatomical information. In this paper, we propose a novel fine-grained segmentation framework termed the “anatomic relation reasoning graph convolutional network” (ARR-GCN), which incorporates prior anatomic relations into the framework learning. In ARR-GCN, a graph is constructed based on the sub-regions to model the class and their relations. Further, to obtain discriminative initial node representations of graph space, a sub-region center module is designed. Most importantly, to explicitly learn the anatomic relations, the prior anatomic-relations among the sub-regions are encoded in the form of an adjacency matrix and embedded into the intermediate node representations to guide framework learning. The ARR-GCN was validated on two FGS-OSA tasks: i) liver segments segmentation, and ii) lung lobes segmentation. Experimental results on both tasks outperformed other state-of-the-art segmentation methods and yielded promising performances by ARR-GCN for suppressing ambiguities among sub-regions.

Precise classification of histopathological images is crucial to computer-aided diagnosis in clinical practice. Magnification-based learning networks have attracted considerable attention for their ability to improve performance in histopathological classification. However, the fusion of pyramids of histopathological images at different magnifications is an under-explored area. In this paper, we proposed a novel deep multi-magnification similarity learning (DSML) approach that can be useful for the interpretation of multi-magnification learning framework and easy to visualize feature representation from low-dimension (e.g., cell-level) to high-dimension (e.g., tissue-level), which has overcome the difficulty of understanding cross-magnification information propagation. It uses a similarity cross entropy loss function designation to simultaneously learn the similarity of the information among cross-magnifications. In order to verify the effectiveness of DMSL, experiments with different network backbones and different magnification combinations were designed, and its ability to interpret was also investigated through visualization. Our experiments were performed on two different histopathological datasets: a clinical nasopharyngeal carcinoma and a public breast cancer BCSS2021 dataset. The results show that our method achieved outstanding performance in classification with a higher value of area under curve, accuracy, and F-score than other comparable methods. Moreover, the reasons behind multi-magnification effectiveness were discussed.

Accurate segmentation of liver blood vessels in magnetic resonance imaging (MRI) images is a challenging task due to the complex tree-like structure and anisotropic diffusion properties of blood vessels. To solve this problem, we propose a new Dual-Path Diffusion Model (DPDM) framework. The framework consists of two collaborative diffusion paths: a local feature learning path based on convolution operations and a global context modeling path based on transform blocks. Local path encodes rich shape priors and preserve spatial details, while global path captures long-distance dependencies and enhance representation. In the decoding phase, the boundary features from the boundary path are fused with the features of the ordinary path decoding, which further enhances the shape sensitivity. In addition, we leverage a multi-task learning scheme to jointly optimize vascular segmentation and boundary prediction tasks in an end-to-end manner. Experiments on retrospective clinical dataset demonstrate that the proposed DPDM framework achieves excellent performance on the liver vessel segmentation task. Compared with state-of-the-art methods, our approach achieved a 6.0% and 7.3% performance improvement in Dice coefficient and IoU index, respectively. Our approach offers a promising solution for automated blood vessel segmentation in precision medicine.

To develop a cross-regional feature fusion model for the classification of cervical tumour subtypes using non-invasive magnetic resonance imaging (MRI), we aimed to explore feature representation in both global and local areas, and to compare their effect on predictive performance.

Visual inspection of hepatocellular carcinoma cancer regions by experienced pathologists in whole-slide images (WSIs) is a challenging, labor-intensive, and time-consuming task because of the large scale and high resolution of WSIs. Therefore, a weakly supervised framework based on a multiscale attention convolutional neural network (MSAN-CNN) was introduced into this process. Herein, patch-based images with image-level normal/tumor annotation (rather than images with pixel-level annotation) were fed into a classification neural network. To further improve the performances of cancer region detection, multiscale attention was introduced into the classification neural network. A total of 100 cases were obtained from The Cancer Genome Atlas and divided into 70 training and 30 testing data sets that were fed into the MSAN-CNN framework. The experimental results showed that this framework significantly outperforms the single-scale detection method according to the area under the curve and accuracy, sensitivity, and specificity metrics. When compared with the diagnoses made by three pathologists, MSAN-CNN performed better than a junior- and an intermediate-level pathologist, and slightly worse than a senior pathologist. Furthermore, MSAN-CNN provided a very fast detection time compared with the pathologists. Therefore, a weakly supervised framework based on MSAN-CNN has great potential to assist pathologists in the fast and accurate detection of cancer regions of hepatocellular carcinoma on WSIs.

The pathologic diagnosis of nasopharyngeal carcinoma (NPC) by different pathologists is often inefficient and inconsistent. We have therefore introduced a deep learning algorithm into this process and compared the performance of the model with that of three pathologists with different levels of experience to demonstrate its clinical value. In this retrospective study, a total of 1970 whole slide images of 731 cases were collected and divided into training, validation, and testing sets. Inception-v3, which is a state-of-the-art convolutional neural network, was trained to classify images into three categories: chronic nasopharyngeal inflammation, lymphoid hyperplasia, and NPC. The mean area under the curve (AUC) of the deep learning model is 0.936 based on the testing set, and its AUCs for the three image categories are 0.905, 0.972, and 0.930, respectively. In the comparison with the three pathologists, the model outperforms the junior and intermediate pathologists, and has only a slightly lower performance than the senior pathologist when considered in terms of accuracy, specificity, sensitivity, AUC, and consistency. To our knowledge, this is the first study about the application of deep learning to NPC pathologic diagnosis. In clinical practice, the deep learning model can potentially assist pathologists by providing a second opinion on their NPC diagnoses.

Volume reconstruction method plays an important role in improving reconstructed volumetric image quality for freehand three-dimensional (3D) ultrasound imaging. By utilizing the capability of programmable graphics processing unit (GPU), we can achieve a real-time incremental volume reconstruction at a speed of 25-50 frames per second (fps). After incremental reconstruction and visualization, hole-filling is performed on GPU to fill remaining empty voxels. However, traditional pixel nearest neighbor–based hole-filling fails to reconstruct volume with high image quality. On the contrary, the kernel regression provides an accurate volume reconstruction method for 3D ultrasound imaging but with the cost of heavy computational complexity. In this paper, a GPU-based fast kernel regression method is proposed for high-quality volume after the incremental reconstruction of freehand ultrasound. The experimental results show that improved image quality for speckle reduction and details preservation can be obtained with the parameter setting of kernel window size of 5×5×5and kernel bandwidth of 1.0. The computational performance of the proposed GPU-based method can be over 200 times faster than that on central processing unit (CPU), and the volume with size of 50 million voxels in our experiment can be reconstructed within 10 seconds.

Ultrasound is one of the most important medical imaging modalities for its real-time and portable imaging advantages. However, the contrast resolution and important details are degraded by the speckle in ultrasound images. Many speckle filtering methods have been developed, but they are suffered from several limitations, difficult to reach a balance between speckle reduction and edge preservation. In this paper, an adaptation of the nonlocal total variation (NLTV) filter is proposed for speckle reduction in ultrasound images. The speckle is modeled via a signal-dependent noise distribution for the log-compressed ultrasound images. Instead of the Euclidian distance, the statistical Pearson distance is introduced in this study for the similarity calculation between image patches via the Bayesian framework. And the Split-Bregman fast algorithm is used to solve the adapted NLTV despeckling functional. Experimental results on synthetic and clinical ultrasound images and comparisons with some classical and recent algorithms are used to demonstrate its improvements in both speckle noise reduction and tissue boundary preservation for ultrasound images.