A transformer-based multi-task deep learning model for simultaneous T-stage identification and segmentation of nasopharyngeal carcinoma.

Background: Accurate tumor target contouring and T staging are vital for precision radiation therapy in nasopharyngeal carcinoma (NPC). Identifying T-stage and contouring the Gross tumor volume (GTV) manually is a laborious and highly time-consuming process. Previous deep learning-based studies have mainly been focused on tumor segmentation, and few studies have specifically addressed the tumor staging of NPC. Objectives: To bridge this gap, we aim to devise a model that can simultaneously identify T-stage and perform accurate segmentation of GTV in NPC. Materials and methods: We have developed a transformer-based multi-task deep learning model that can perform two tasks simultaneously: delineating the tumor contour and identifying T-stage. Our retrospective study involved contrast-enhanced T1-weighted images (CE-T1WI) of 320 NPC patients (T-stage: T1-T4) collected between 2017 and 2020 at our institution, which were randomly allocated into three cohorts for three-fold cross-validations, and conducted the external validation using an independent test set. We evaluated the predictive performance using the area under the receiver operating characteristic curve (ROC-AUC) and accuracy (ACC), with a 95% confidence interval (CI), and the contouring performance using the Dice similarity coefficient (DSC) and average surface distance (ASD). Results: Our multi-task model exhibited sound performance in GTV contouring (median DSC: 0.74; ASD: 0.97 mm) and T staging (AUC: 0.85, 95% CI: 0.82-0.87) across 320 patients. In early T category tumors, the model achieved a median DSC of 0.74 and ASD of 0.98 mm, while in advanced T category tumors, it reached a median DSC of 0.74 and ASD of 0.96 mm. The accuracy of automated T staging was 76% (126 of 166) for early stages (T1-T2) and 64% (99 of 154) for advanced stages (T3-T4). Moreover, experimental results show that our multi-task model outperformed the other single-task models. Conclusions: This study emphasized the potential of multi-task model for simultaneously delineating the tumor contour and identifying T-stage. The multi-task model harnesses the synergy between these interrelated learning tasks, leading to improvements in the performance of both tasks. The performance demonstrates the potential of our work for delineating the tumor contour and identifying T-stage and suggests that it can be a practical tool for supporting clinical precision radiation therapy.

Cross-Domain Mutual-Assistance Learning Framework for Fully Automated Diagnosis of Primary Tumor in Nasopharyngeal Carcinoma.

Accurate T-staging of nasopharyngeal carcinoma (NPC) holds paramount importance in guiding treatment decisions and prognosticating outcomes for distinct risk groups. Regrettably, the landscape of deep learning-based techniques for T-staging in NPC remains sparse, and existing methodologies often exhibit suboptimal performance due to their neglect of crucial domain-specific knowledge pertinent to primary tumor diagnosis. To address these issues, we propose a new cross-domain mutual-assistance learning framework for fully automated diagnosis of primary tumor using H&N MR images. Specifically, we tackle primary tumor diagnosis task with the convolutional neural network consisting of a 3D cross-domain knowledge perception network (CKP net) for excavated cross-domain-invariant features emphasizing tumor intensity variations and internal tumor heterogeneity, and a multi-domain mutual-information sharing fusion network (M2SF net), comprising a dual-pathway domain-specific representation module and a mutual information fusion module, for intelligently gauging and amalgamating multi-domain, multi-scale T-stage diagnosis-oriented features. The proposed 3D cross-domain mutual-assistance learning framework not only embraces task-specific multi-domain diagnostic knowledge but also automates the entire process of primary tumor diagnosis. We evaluate our model on an internal and an external MR images dataset in a three-fold cross-validation paradigm. Exhaustive experimental results demonstrate that our method outperforms the state-of-the-art algorithms, and obtains promising performance for tumor segmentation and T-staging. These findings underscore its potential for clinical application, offering valuable assistance to clinicians in treatment decision-making and prognostication for various risk groups.

Aqueous Processable One-Dimensional Polypyrrole Nanostructured by Lignocellulose Nanofibril: A Conductive Interfacing Biomaterial.

One-dimensional (1D) nanomaterials of conductive polypyrrole (PPy) are competitive biomaterials for constructing bioelectronics to interface with biological systems. Synergistic synthesis using lignocellulose nanofibrils (LCNF) as a structural template in chemical oxidation of pyrrole with Fe(III) ions facilitates surface-confined polymerization of pyrrole on the nanofibril surface within a submicrometer- and micrometer-scale fibril length. It yields a core-shell nanocomposite of PPy@LCNF, wherein the surface of each individual fibril is coated with a thin nanoscale layer of PPy. A highly positive surface charge originating from protonated PPy gives this 1D nanomaterial a durable aqueous dispersity. The fibril-fibril entanglement in the PPy@LCNFs facilely supported versatile downstream processing, e.g., spray thin-coating on glass, flexible membranes with robust mechanics, or three-dimensional cryogels. A high electrical conductivity in the magnitude of several to 12 S·cm-1 was confirmed for the solid-form PPy@LCNFs. The PPy@LCNFs are electroactive and show potential cycling capacity, encompassing a large capacitance. Dynamic control of the doping/undoping process by applying an electric field combines electronic and ionic conductivity through the PPy@LCNFs. The low cytotoxicity of the material is confirmed in noncontact cell culture of human dermal fibroblasts. This study underpins the promises for this nanocomposite PPy@LCNF as a smart platform nanomaterial in constructing interfacing bioelectronics.

PHF14 enhances DNA methylation of SMAD7 gene to promote TGF-ß-driven lung adenocarcinoma metastasis.

Aberrant activation of TGF-ß signaling plays a pivotal role in cancer metastasis and progression. However, molecular mechanisms underlying the dysregulation of TGF-ß pathway remain to be understood. Here, we found that SMAD7, a direct downstream transcriptional target and also a key antagonist of TGF-ß signaling, is transcriptionally suppressed in lung adenocarcinoma (LAD) due to DNA hypermethylation. We further identified that PHF14 binds DNMT3B and serves as a DNA CpG motif reader, recruiting DNMT3B to the SMAD7 gene locus, resulting in DNA methylation and transcriptional suppression of SMAD7. Our in vitro and in vivo experiments showed that PHF14 promotes metastasis through binding DNMT3B to suppress SMAD7 expression. Moreover, our data revealed that PHF14 expression correlates with lowered SMAD7 level and shorter survival of LAD patients, and importantly that SMAD7 methylation level of circulating tumor DNA (ctDNA) can potentially be used for prognosis prediction. Together, our present study illustrates a new epigenetic mechanism, mediated by PHF14 and DNMT3B, in the regulation of SMAD7 transcription and TGF-ß-driven LAD metastasis, and suggests potential opportunities for LAD prognosis.

A multi-perspective information aggregation network for automatedT-staging detection of nasopharyngeal carcinoma.

AccurateT-staging is important when planning personalized radiotherapy. However,T-staging via manual slice-by-slice inspection is time-consuming while tumor sizes and shapes are heterogeneous, and junior physicians find such inspection challenging. With inspiration from oncological diagnostics, we developed a multi-perspective aggregation network that incorporated various diagnosis-oriented knowledge which allowed automated nasopharyngeal carcinomaT-staging detection (TSD Net). Specifically, our TSD Net was designed in multi-branch architecture, which can capture tumor size and shape information (basic knowledge), strongly correlated contextual features, and associations between the tumor and surrounding tissues. We defined the association between the tumor and surrounding tissues by a signed distance map which can embed points and tumor contours in higher-dimensional spaces, yielding valuable information regarding the locations of tissue associations. TSD Net finally outputs aT1-T4 stage prediction by aggregating data from the three branches. We evaluated TSD Net by using the T1-weighted contrast-enhanced magnetic resonance imaging database of 320 patients in a three-fold cross-validation manner. The results show that the proposed method achieves a mean area under the curve (AUC) as high as 87.95%. We also compared our method to traditional classifiers and a deep learning-based method. Our TSD Net is efficient and accurate and outperforms other methods.

Graph-enhanced U-Net for semi-supervised segmentation of pancreas from abdomen CT scan.

Objective. Accurate segmentation of the pancreas from abdomen CT scans is highly desired for diagnosis and treatment follow-up of pancreatic diseases. However, the task is challenged by large anatomical variations, low soft-tissue contrast, and the difficulty in acquiring a large set of annotated volumetric images for training. To overcome these problems, we propose a new segmentation network and a semi-supervised learning framework to alleviate the lack of annotated images and improve the accuracy of segmentation.Approach.In this paper, we propose a novel graph-enhanced pancreas segmentation network (GEPS-Net), and incorporate it into a semi-supervised learning framework based on iterative uncertainty-guided pseudo-label refinement. Our GEPS-Net plugs a graph enhancement module on top of the CNN-based U-Net to focus on the spatial relationship information. For semi-supervised learning, we introduce an iterative uncertainty-guided refinement process to update pseudo labels by removing low-quality and incorrect regions.Main results.Our method was evaluated by a public dataset with four-fold cross-validation and achieved the DC of 84.22%, improving 5.78% compared to the baseline. Further, the overall performance of our proposed method was the best compared with other semi-supervised methods trained with only 6 or 12 labeled volumes.Significance.The proposed method improved the segmentation performance of the pancreas in CT images under the semi-supervised setting. It will assist doctors in early screening and making accurate diagnoses as well as adaptive radiotherapy.

Low-dose cadmium exposure promotes osteoclastogenesis by enhancing autophagy via inhibiting the mTOR/p70S6K1 signaling pathway.

Cadmium (Cd)-induced bone damage may be mediated through activating osteoclastogenesis. However, the underlying mechanism is unknown. The purpose of this study was to explore the effect and possible mechanism of CdCl2-induced osteoclastogenesis in RAW264.7 cells. We found that a low concentration of CdCl2 (0.025 and 0.050 µM) did not affect the viability of RAW264.7 cells, but promoted osteoclastogenesis. A low concentration of CdCl2 increased the mRNA and protein expression of osteoclastogenesis-related genes. TRAP staining and transmission electron microscopy (TEM) also demonstrated that CdCl2 promoted osteoclastogenesis. A low concentration of CdCl2 upregulated the levels of LC3-II and Beclin-1, and decreased p62 expression. TEM showed relatively abundant autophagic vacuoles (autophagosomes) after CdCl2 exposure. A low concentration of CdCl2 downregulated the expression levels of Mtor and p70S6K1, and the relative protein expression ratios of p-mTOR/mTOR and p-p70S6K1/p70S6K1. When cells were treated with the autophagy inhibitor chloroquine (CQ) or mTOR activator MHY1485 combined with CdCl2, the expressions of osteoclastogenesis related-genes were decreased and autophagy was attenuated compared with cells treated with CdCl2 alone. Deficiencies in autophagosomes and osteoclasts were also observed. Taken together, the results indicate that a low concentration of CdCl2 promotes osteoclastogenesis by enhancing autophagy via inhibiting the mTOR/p70S6K1 signaling pathway.

The Prognostic Impact of Lymph Node Dissection on Primary Tumor Resection for Stage IV Non-Small Cell Lung Cancer: A Population-Based Study.

Objective: Selected patients with stage IV non-small cell lung cancer (NSCLC) who underwent primary tumor resection have witnessed a survival benefit. Whether additional lymph node dissection (LND) would result in a better effect remain unknown. We investigated the prognostic impact of LND on patients with stage IV NSCLC who received primary tumor resection (PTR). Methods: Patients with stage IV NSCLC who underwent PTR were identified from the Surveillance, Epidemiology, and End Results database from 2004 to 2016. Propensity-score matching was performed to minimize the confounding effect, and lung cancer-specific survival (CSS) and overall survival (OS) were compared after matching. Multivariable Cox regression was used to identify prognostic factors and to adjust for covariates in subgroup analysis. The effect of the number of lymph nodes examined on the CSS was evaluated by repeating the Cox analysis in a binary method. Results: A total of 4,114 patients with stage IV NSCLC who receive surgery met our criteria, of which 2,622 (63.73%) underwent LND and 628 patients were identified 1:1 in LND and non-LND groups after matching. Compared with the non-LND group, the LND group had a longer CSS (median: 23 vs. 16 months, p < 0.001) and OS (median: 21 vs. 15 months, p < 0.001). Multivariable regression showed that LND was independently associated with favorable CCS [hazard ratio (HR) = 0.78, 95% confidence interval (CI) 0.69-0.89, P < 0.001] and OS (HR = 0.79, 95% CI 0.70-0.89, P < 0.001). Subgroup analysis suggested that LND is an independent favorable predictor to survival in the surgical patients who were older age (>60 years old), female, T3-4, N0, and M1a stage and those who underwent sublobar resection. In addition, a statistically significant CCS benefit was associated with an increasing number of lymph nodes examined through 25 lymph nodes. Conclusions: LND with a certain range of lymph nodes number examined was associated with improved survival for patients with stage IV NSCLC who received primary tumor resection. The results may have implications for guidelines on lymph nodes management in selective advanced NSCLC for surgery.

CCT5 induces epithelial-mesenchymal transition to promote gastric cancer lymph node metastasis by activating the Wnt/ß-catenin signalling pathway.

BACKGROUND: Lymph node (LN) metastasis confers gastric cancer (GC) progression, poor survival and cancer-related death. Aberrant activation of Wnt/ß-catenin promotes epithelial-mesenchymal transition (EMT) and LN metastasis, whereas the constitutive activation mutation of Wnt/ß-catenin is rare in GC, suggesting that the underlying mechanisms enhancing Wnt/ß-catenin activation need to be further investigated and understood. METHODS: Bioinformatics analyses and immunohistochemistry (IHC) were used to identify and detect LN metastasis-related genes in GC. Cellular functional assays and footpad inoculation mouse model illustrate the biological function of CCT5. Co-immunoprecipitation assays, western blot and qPCR elucidate the interaction between CCT5 and E-cadherin, and the regulation on ß-catenin activity. RESULTS: CCT5 is upregulated in LN metastatic GCs and correlates with poor prognosis. In vitro assays prove that CCT5 markedly promotes GC cell proliferation, anti-anoikis, invasion and lymphatic tube formation. Moreover, CCT5 enhances xenograft GC growth and popliteal lymph node metastasis in vivo. Furthermore, CCT5 binds the cytoplasmic domain of E-cadherin and abrogates the interaction between E-cadherin and ß-catenin, thereby releasing ß-catenin to the nucleus and enhancing Wnt/ß-catenin signalling activity and EMT. CONCLUSION: CCT5 promotes GC progression and LN metastasis by enhancing wnt/ß-catenin activation, suggesting a great potential of CCT5 as a biomarker for GC diagnosis and therapy.

An RFC4/Notch1 signaling feedback loop promotes NSCLC metastasis and stemness.

Notch signaling represents a key mechanism mediating cancer metastasis and stemness. To understand how Notch signaling is overactivated to couple tumor metastasis and self-renewal in NSCLC cells, we performed the current study and showed that RFC4, a DNA replication factor amplified in more than 40% of NSCLC tissues, directly binds to the Notch1 intracellular domain (NICD1) to competitively abrogate CDK8/FBXW7-mediated degradation of NICD1. Moreover, RFC4 is a functional transcriptional target gene of Notch1 signaling, forming a positive feedback loop between high RFC4 and NICD1 levels and sustained overactivation of Notch signaling, which not only leads to NSCLC tumorigenicity and metastasis but also confers NSCLC cell resistance to treatment with the clinically tested drug DAPT against NICD1 synthesis. Furthermore, together with our study, analysis of two public datasets involving more than 1500 NSCLC patients showed that RFC4 gene amplification, and high RFC4 and NICD1 levels were tightly correlated with NSCLC metastasis, progression and poor patient prognosis. Therefore, our study characterizes the pivotal roles of the positive feedback loop between RFC4 and NICD1 in coupling NSCLC metastasis and stemness properties and suggests its therapeutic and diagnostic/prognostic potential for NSCLC therapy.

Breast ultrasound image segmentation: A coarse-to-fine fusion convolutional neural network.

PURPOSE: Breast ultrasound (BUS) image segmentation plays a crucial role in computer-aided diagnosis systems for BUS examination, which are useful for improved accuracy of breast cancer diagnosis. However, such performance remains a challenging task owing to the poor image quality and large variations in the sizes, shapes, and locations of breast lesions. In this paper, we propose a new convolutional neural network with coarse-to-fine feature fusion to address the aforementioned challenges. METHODS: The proposed fusion network consists of an encoder path, a decoder path, and a core fusion stream path (FSP). The encoder path is used to capture the context information, and the decoder path is used for localization prediction. The FSP is designed to generate beneficial aggregate feature representations (i.e., various-sized lesion features, aggregated coarse-to-fine information, and high-resolution edge characteristics) from the encoder and decoder paths, which are eventually used for accurate breast lesion segmentation. To better retain the boundary information and alleviate the effect of image noise, we input the superpixel image along with the original image to the fusion network. Furthermore, a weighted-balanced loss function was designed to address the problem of lesion regions having different sizes. We then conducted exhaustive experiments on three public BUS datasets to evaluate the proposed network. RESULTS: The proposed method outperformed state-of-the-art (SOTA) segmentation methods on the three public BUS datasets, with average dice similarity coefficients of 84.71(±1.07), 83.76(±0.83), and 86.52(±1.52), average intersection-over-union values of 76.34(±1.50), 75.70(±0.98), and 77.86(±2.07), average sensitivities of 86.66(±1.82), 85.21(±1.98), and 87.21(±2.51), average specificities of 97.92(±0.46), 98.57(±0.19), and 99.42(±0.21), and average accuracies of 95.89(±0.57), 97.17(±0.3), and 98.51(±0.3). CONCLUSIONS: The proposed fusion network could effectively segment lesions from BUS images, thereby presenting a new feature fusion strategy to handle challenging task of segmentation, while outperforming the SOTA segmentation methods. The code is publicly available at https://github.com/mniwk/CF2-NET.

AKT-induced lncRNA VAL promotes EMT-independent metastasis through diminishing Trim16-dependent Vimentin degradation.

Despite the importance of AKT overactivation in tumor progression, results from clinical trials of various AKT inhibitors remain suboptimal, suggesting that AKT-driven tumor metastasis needs to be further understood. Herein, based on long non-coding RNA (lncRNA) profiling induced by active AKT, we identify that VAL (Vimentin associated lncRNA, LINC01546), which is directly induced by AKT/STAT3 signaling, functions as a potent pro-metastatic molecule and is essential for active AKT-induced tumor invasion, metastasis and anoikis resistance in lung adenocarcinoma (LAD). Impressively, chemosynthetic siRNAs against VAL shows great therapeutic potential in AKT overactivation-driven metastasis. Interestingly, similar to activated AKT in LAD cells, although unable to induce epithelial-mesenchymal transition (EMT), VAL exerts potent pro-invasive and pro-metastatic effects through directly binding to Vimentin and competitively abrogating Trim16-depedent Vimentin polyubiquitination and degradation. Taken together, our study provides an interesting demonstration of a lncRNA-mediated mechanism for active AKT-driven EMT-independent LAD metastasis and indicates the great potential of targeting VAL or Vimentin stability as a therapeutic approach.

Multi-View Spatial Aggregation Framework for Joint Localization and Segmentation of Organs at Risk in Head and Neck CT Images.

Accurate segmentation of organs at risk (OARs) from head and neck (H&N) CT images is crucial for effective H&N cancer radiotherapy. However, the existing deep learning methods are often not trained in an end-to-end fashion, i.e., they independently predetermine the regions of target organs before organ segmentation, causing limited information sharing between related tasks and thus leading to suboptimal segmentation results. Furthermore, when conventional segmentation network is used to segment all the OARs simultaneously, the results often favor big OARs over small OARs. Thus, the existing methods often train a specific model for each OAR, ignoring the correlation between different segmentation tasks. To address these issues, we propose a new multi-view spatial aggregation framework for joint localization and segmentation of multiple OARs using H&N CT images. The core of our framework is a proposed region-of-interest (ROI)-based fine-grained representation convolutional neural network (CNN), which is used to generate multi-OAR probability maps from each 2D view (i.e., axial, coronal, and sagittal view) of CT images. Specifically, our ROI-based fine-grained representation CNN (1) unifies the OARs localization and segmentation tasks and trains them in an end-to-end fashion, and (2) improves the segmentation results of various-sized OARs via a novel ROI-based fine-grained representation. Our multi-view spatial aggregation framework then spatially aggregates and assembles the generated multi-view multi-OAR probability maps to segment all the OARs simultaneously. We evaluate our framework using two sets of H&N CT images and achieve competitive and highly robust segmentation performance for OARs of various sizes.

Postoperative glioma segmentation in CT image using deep feature fusion model guided by multi-sequence MRIs.

OBJECTIVES: Computed tomography (CT) and magnetic resonance imaging (MRI) are the most commonly selected methods for imaging gliomas. Clinically, radiotherapists always delineate the CT glioma region with reference to multi-modal MR image information. On this basis, we develop a deep feature fusion model (DFFM) guided by multi-sequence MRIs for postoperative glioma segmentation in CT images. METHODS: DFFM is a multi-sequence MRI-guided convolutional neural network (CNN) that iteratively learns the deep features from CT images and multi-sequence MR images simultaneously by utilizing a multi-channel CNN architecture, and then combines these two deep features together to produce the segmentation result. The whole network is optimized together via a standard back-propagation. A total of 59 CT and MRI datasets (T1/T2-weighted FLAIR, T1-weighted contrast-enhanced, T2-weighted) of postoperative gliomas as tumor grade II (n = 24), grade III (n = 18), or grade IV (n = 17) were included. Dice coefficient (DSC), precision, and recall were used to measure the overlap between automated segmentation results and manual segmentation. The Wilcoxon signed-rank test was used for statistical analysis. RESULTS: DFFM showed a significantly (p < 0.01) higher DSC of 0.836 than U-Net trained by single CT images and U-Net trained by stacking the CT and multi-sequence MR images, which yielded 0.713 DSC and 0.818 DSC, respectively. The precision values showed similar behavior as DSC. Moreover, DSC and precision values have no significant statistical difference (p > 0.01) with difference grades. CONCLUSIONS: DFFM enables the accurate automated segmentation of CT postoperative gliomas of profit guided by multi-sequence MR images and may thus improve and facilitate radiotherapy planning. KEY POINTS: • A fully automated deep learning method was developed to segment postoperative gliomas on CT images guided by multi-sequence MRIs. • CT and multi-sequence MR image integration allows for improvements in deep learning postoperative glioma segmentation method. • This deep feature fusion model produces reliable segmentation results and could be useful in delineating GTV in postoperative glioma radiotherapy planning.

Incorporation of Complexation into a Coamorphous System Dramatically Enhances Dissolution and Eliminates Gelation of Amorphous Lurasidone Hydrochloride.

As a BCS II drug, the atypical antipsychotic agent lurasidone hydrochloride (LH) has low oral bioavailability mainly because of its poor aqueous solubility/dissolution. Unexpectedly, amorphous LH exhibited a much lower dissolution than that of its stable crystalline form arising from its gelation during the dissolution process. In the current study, a supramolecular coamorphous system of LH with l-cysteine hydrochloride (CYS) was prepared and characterized by powder X-ray diffraction and differential scanning calorimetry. Surprisingly, in comparison to crystalline and amorphous LH, such a coamorphous system dramatically enhanced solubility (at least ∼50-fold in the physiological pH range) and dissolution (∼1200-fold) of LH, and exhibited superior physical stability under long-term storage condition. More importantly, the coamorphous system was able to eliminate gelation of amorphous LH during dissolution. In order to further explore the mechanism of such improvement, the internal interactions of the coamorphous system in the solid state and in aqueous solution were investigated. Fourier transform infrared spectroscopy, Raman spectroscopy, and solid-state 13C NMR suggested that intermolecular hydrogen bonds formed between the nitrogen atom in the benzisothiazole ring of LH and the NH3+ group of CYS after coamorphization. A fluorescence quenching test with a Stern-Volmer plot and density functional theory modeling, phase-solubility study, and NMR test in D2O indicated that ground-state complexation occurred between LH and CYS in aqueous solution, which contributed to the solubility and dissolution enhancement of LH. The current study offers a promising strategy to overcome poor solubility/dissolution and be able to eliminate gelation of amorphous materials by coamorphization and complexation.

Boosting-based Cascaded Convolutional Neural Networks for the Segmentation of CT Organs-at-risk in Nasopharyngeal Carcinoma.

PURPOSE: Accurately segmenting organs-at-risk (OARs) is a key step in the effective planning of radiation therapy for nasopharyngeal carcinoma (NPC) treatment. In OAR segmentation of the head and neck CT, the low contrast and surrounding adhesion tissues of the parotids, thyroids, and optic nerves result in the difficulty in segmentation and lower accuracy of automatic segmentation for these organs than the other organs. In this paper, we propose a cascaded network structure to delineate these three OARs for NPC radiotherapy by combining deep learning and Boosting algorithm. MATERIALS AND METHODS: The CT images of 140 NPC patients treated with radiotherapy were collected, and each of the three OAR annotations was respectively delineated by an experienced rater and reviewed by a professional radiologist (with 10 years of experience). The datasets (140 patients) were divided into a training set (100 patients), a validation set (20 patients), and a test set (20 patients). From the Boosting method for combining multiple classifiers, three cascaded CNNs for segmentation were combined. The first network was trained with the traditional approach. The second one was trained on patterns (pixels) filtered by the first net. That is, the second machine recognized a mix of patterns (pixels), 50% of which was accurately identified by the first net. Finally, the third net was trained on the new patterns (pixels) screened jointly by the first and second networks. During the test, the outputs of the three nets were considered to obtain the final output. Dice similarity coefficient (DSC), 95th percentile of the Hausdorff distance (95% HD), and volume overlap error (VOE) were used to assess the method performance. RESULTS: The mean DSC (%) values were above 92.26 for the parotids, above 92.29 for the thyroids, and above 89.37 for the optic nerves. The mean 95% HDs (mm) were approximately 3.08 for the parotids, 2.64 for the thyroids, and 2.03 for the optic nerves. The mean VOE (%) values were approximately 14.16 for the parotids, 14.94 for the thyroids, and 19.07 for the optic nerves. CONCLUSION: The proposed cascaded deep learning structure could achieve high performance compared with existing single-network or other segmentation algorithms.

Automated detection and classification of thyroid nodules in ultrasound images using clinical-knowledge-guided convolutional neural networks.

Accurate diagnosis of thyroid nodules using ultrasonography is a valuable but tough task even for experienced radiologists, considering both benign and malignant nodules have heterogeneous appearances. Computer-aided diagnosis (CAD) methods could potentially provide objective suggestions to assist radiologists. However, the performance of existing learning-based approaches is still limited, for direct application of general learning models often ignores critical domain knowledge related to the specific nodule diagnosis. In this study, we propose a novel deep-learning-based CAD system, guided by task-specific prior knowledge, for automated nodule detection and classification in ultrasound images. Our proposed CAD system consists of two stages. First, a multi-scale region-based detection network is designed to learn pyramidal features for detecting nodules at different feature scales. The region proposals are constrained by the prior knowledge about size and shape distributions of real nodules. Then, a multi-branch classification network is proposed to integrate multi-view diagnosis-oriented features, in which each network branch captures and enhances one specific group of characteristics that were generally used by radiologists. We evaluated and compared our method with the state-of-the-art CAD methods and experienced radiologists on two datasets, i.e. Dataset I and Dataset II. The detection and diagnostic accuracy on Dataset I were 97.5% and 97.1%, respectively. Besides, our CAD system also achieved better performance than experienced radiologists on Dataset II, with improvements of accuracy for 8%. The experimental results demonstrate that our proposed method is effective in the discrimination of thyroid nodules.

Deep-learning-based detection and segmentation of organs at risk in nasopharyngeal carcinoma computed tomographic images for radiotherapy planning.

OBJECTIVE: Accurate detection and segmentation of organs at risks (OARs) in CT image is the key step for efficient planning of radiation therapy for nasopharyngeal carcinoma (NPC) treatment. We develop a fully automated deep-learning-based method (termed organs-at-risk detection and segmentation network (ODS net)) on CT images and investigate ODS net performance in automated detection and segmentation of OARs. METHODS: The ODS net consists of two convolutional neural networks (CNNs). The first CNN proposes organ bounding boxes along with their scores, and then a second CNN utilizes the proposed bounding boxes to predict segmentation masks for each organ. A total of 185 subjects were included in this study for statistical comparison. Sensitivity and specificity were performed to determine the performance of the detection and the Dice coefficient was used to quantitatively measure the overlap between automated segmentation results and manual segmentation. Paired samples t tests and analysis of variance were employed for statistical analysis. RESULTS: ODS net provides an accurate detection result with a sensitivity of 0.997 to 1 for most organs and a specificity of 0.983 to 0.999. Furthermore, segmentation results from ODS net correlated strongly with manual segmentation with a Dice coefficient of more than 0.85 in most organs. A significantly higher Dice coefficient for all organs together (p = 0.0003 < 0.01) was obtained in ODS net (0.861 ± 0.07) than in fully convolutional neural network (FCN) (0.8 ± 0.07). The Dice coefficients of each OAR did not differ significantly between different T-staging patients. CONCLUSION: The ODS net yielded accurate automated detection and segmentation of OARs in CT images and thereby may improve and facilitate radiotherapy planning for NPC. KEY POINTS: • A fully automated deep-learning method (ODS net) is developed to detect and segment OARs in clinical CT images. • This deep-learning-based framework produces reliable detection and segmentation results and thus can be useful in delineating OARs in NPC radiotherapy planning. • This deep-learning-based framework delineating a single image requires approximately 30 s, which is suitable for clinical workflows.

Concomitant pulmonary and thyroid tumors identified by FDG PET/CT and immunohistochemical techniques.

BACKGROUND: The exact diagnosis of double primary papillary adenocarcinoma of thyroid and lung is even rarer, to our knowledge no report in the literature by [¹8F]-2-fluoro-2-deoxy-D-glucose-positron emission tomography/X-ray CT(FDG PET/CT) with surgical specimens immunohistochemistry(IHC). We report a patient with abnormal FDG PET/CT in thyroid and lung, this unusual presentation may lead to misdiagnosis without surgical specimens IHC. CASE PRESENTATION: A 56-year-old man with coughing three months. FDG PET/CT was performed, and resection specimens of lung and thyroid were detected by hematoxylin eosin staining (HE) and IHC. PET/CT: lung tumor SUVmax: 3.69, delay: 5.17; and thyroid tumor SUVmax 19.97. HE reveal papillary adenocarcinoma, but histological differentiation of primary pulmonary adenocarcinoma from metastatic adenocarcinoma is sometimes difficult because of their phenotypic similarities. So IHC was performed, the IHC of lung tumor: cytokeratin 20 (CK20)⁻, thyroglobulin(Tg)⁻, cytokeratin7(CK7)+, thyroid transcription factor-1 (TTF-1)+; thyroid tumor: CK7+, TTF-1+, thyroglobulin+, CK20⁻. Therefore, the final diagnosis was double primary adenocarcinomas of thyroid and lung. CONCLUSION: FDG PET/CT has preliminary diagnostic capacity of multiple primary tumors; the final diagnosis should be adopted for specimens after tumor-specific markers IHC to obtain. Consequently, effective therapeutic approaches can be designed and conducted.

[Up-regulation of Fas is related to apoptosis of HeLa cells induced by co-immobilized TNF-alpha/IFN-gamma].

This study was aimed to examine the expression of apoptosis-associated gene Fas in HeLa cell, explore the effects of the co-immobilized cytokines (tumor necrosis factor-alpha and interferon-gamma), and probe the potential mechanism of action. The preparation and application of the research couple IFN-gamma and TNF-alpha to the polystyrene cell culture plate were performed using the Photo-immobilization method, with different doses (20 ng/well and 200 ng/well) and synthesized optical active material. HeLa cells were treated with cytokines for two dose and 1, 3, 6 days. The result showed that the free cytokines induced HeLa apoptosis quickly, yet the HeLa apoptosis induced by co-immobilized cytokines had longer effect.