Multicenter radio-multiomic analysis for predicting breast cancer outcome and unravelling imaging-biological connection.

Radiomics offers a noninvasive avenue for predicting clinicopathological factors. However, thorough investigations into a robust breast cancer outcome-predicting model and its biological significance remain limited. This study develops a robust radiomic model for prognosis prediction, and further excavates its biological foundation and transferring prediction performance. We retrospectively collected preoperative dynamic contrast-enhanced MRI data from three distinct breast cancer patient cohorts. In FUSCC cohort (n = 466), Lasso was used to select features correlated with patient prognosis and multivariate Cox regression was utilized to integrate these features and build the radiomic risk model, while multiomic analysis was conducted to investigate the model's biological implications. DUKE cohort (n = 619) and I-SPY1 cohort (n = 128) were used to test the performance of the radiomic signature in outcome prediction. A thirteen-feature radiomic signature was identified in the FUSCC cohort training set and validated in the FUSCC cohort testing set, DUKE cohort and I-SPY1 cohort for predicting relapse-free survival (RFS) and overall survival (OS) (RFS: p = 0.013, p = 0.024 and p = 0.035; OS: p = 0.036, p = 0.005 and p = 0.027 in the three cohorts). Multiomic analysis uncovered metabolic dysregulation underlying the radiomic signature (ATP metabolic process: NES = 1.84, p-adjust = 0.02; cholesterol biosynthesis: NES = 1.79, p-adjust = 0.01). Regarding the therapeutic implications, the radiomic signature exhibited value when combining clinical factors for predicting the pathological complete response to neoadjuvant chemotherapy (DUKE cohort, AUC = 0.72; I-SPY1 cohort, AUC = 0.73). In conclusion, our study identified a breast cancer outcome-predicting radiomic signature in a multicenter radio-multiomic study, along with its correlations with multiomic features in prognostic risk assessment, laying the groundwork for future prospective clinical trials in personalized risk stratification and precision therapy.

A preoperative radiogenomic model based on quantitative heterogeneity for predicting outcomes in triple-negative breast cancer patients who underwent neoadjuvant chemotherapy.

BACKGROUND: Triple-negative breast cancer (TNBC) is highly heterogeneous, resulting in different responses to neoadjuvant chemotherapy (NAC) and prognoses among patients. This study sought to characterize the heterogeneity of TNBC on MRI and develop a radiogenomic model for predicting both pathological complete response (pCR) and prognosis. MATERIALS AND METHODS: In this retrospective study, TNBC patients who underwent neoadjuvant chemotherapy at Fudan University Shanghai Cancer Center were enrolled as the radiomic development cohort (n = 315); among these patients, those whose genetic data were available were enrolled as the radiogenomic development cohort (n = 98). The study population of the two cohorts was randomly divided into a training set and a validation set at a ratio of 7:3. The external validation cohort (n = 77) included patients from the DUKE and I-SPY 1 databases. Spatial heterogeneity was characterized using features from the intratumoral subregions and peritumoral region. Hemodynamic heterogeneity was characterized by kinetic features from the tumor body. Three radiomics models were developed by logistic regression after selecting features. Model 1 included subregional and peritumoral features, Model 2 included kinetic features, and Model 3 integrated the features of Model 1 and Model 2. Two fusion models were developed by further integrating pathological and genomic features (PRM: pathology-radiomics model; GPRM: genomics-pathology-radiomics model). Model performance was assessed with the AUC and decision curve analysis. Prognostic implications were assessed with Kaplan‒Meier curves and multivariate Cox regression. RESULTS: Among the radiomic models, the multiregional model representing multiscale heterogeneity (Model 3) exhibited better pCR prediction, with AUCs of 0.87, 0.79, and 0.78 in the training, internal validation, and external validation sets, respectively. The GPRM showed the best performance for predicting pCR in the training (AUC = 0.97, P = 0.015) and validation sets (AUC = 0.93, P = 0.019). Model 3, PRM and GPRM could stratify patients by disease-free survival, and a predicted nonpCR was associated with poor prognosis (P = 0.034, 0.001 and 0.019, respectively). CONCLUSION: Multiscale heterogeneity characterized by DCE-MRI could effectively predict the pCR and prognosis of TNBC patients. The radiogenomic model could serve as a valuable biomarker to improve the prediction performance.

LncRNA IRAIN overcomes imatinib resistance in chronic myeloid leukemia via NF-κB/CD44 pathway inhibition.

The development of tyrosine kinase inhibitors (TKIs) has revolutionarily increased the overall survival of patients with chronic myeloid leukemia (CML). However, drug resistance remains a major obstacle. Here, we demonstrated that a BCR-ABL1-independent long non-coding RNA, IRAIN, is constitutively expressed at low levels in CML, resulting in imatinib resistance. IRAIN knockdown decreased the sensitivity of CD34+ CML blasts and cell lines to imatinib, whereas IRAIN overexpression significantly increased sensitivity. Mechanistically, IRAIN downregulates CD44, a membrane receptor favorably affecting TKI resistance, by binding to the nuclear factor kappa B subunit p65 to reduce the expression of p65 and phosphorylated p65. Therefore, the demethylating drug decitabine, which upregulates IRAIN, combined with imatinib, formed a dual therapy strategy which can be applied to CML with resistance to TKIs.

Radiotherapy-Induced Astrocyte Senescence Promotes an Immunosuppressive Microenvironment in Glioblastoma to Facilitate Tumor Regrowth.

Accumulating evidence suggests that changes in the tumor microenvironment caused by radiotherapy are closely related to the recurrence of glioma. However, the mechanisms by which such radiation-induced changes are involved in tumor regrowth have not yet been fully investigated. In the present study, how cranial irradiation-induced senescence in non-neoplastic brain cells contributes to glioma progression is explored. It is observed that senescent brain cells facilitated tumor regrowth by enhancing the peripheral recruitment of myeloid inflammatory cells in glioblastoma. Further, it is identified that astrocytes are one of the most susceptible senescent populations and that they promoted chemokine secretion in glioma cells via the senescence-associated secretory phenotype. By using senolytic agents after radiotherapy to eliminate these senescent cells substantially prolonged survival time in preclinical models. The findings suggest the tumor-promoting role of senescent astrocytes in the irradiated glioma microenvironment and emphasize the translational relevance of senolytic agents for enhancing the efficacy of radiotherapy in gliomas.

Components of the tumor immune microenvironment based on m-IHC correlate with prognosis and subtype of triple-negative breast cancer.

BACKGROUND AND AIM: The spatial distribution and interactions of cells in the tumor immune microenvironment (TIME) might be related to the different responses of triple-negative breast cancer (TNBC) to immunomodulators. The potential of multiplex IHC (m-IHC) in evaluating the TIME has been reported, but the efficacy is insufficient. We aimed to research whether m-IHC results could be used to reflect the TIME, and thus to predict prognosis and complement the TNBC subtyping system. METHODS: The clinical, imaging, and prognosis data for 86 TNBC patients were retrospectively reviewed. CD3, CD4, CD8, Foxp3, PD-L1, and Pan-CK markers were stained by m-IHC. Particular cell spatial distributions and interactions in the TIME were evaluated with the HALO multispectral analysis platform. Then, we calculated the prognostic value of components of the TIME and their correlations with TNBC transcriptomic subtypes and MRI radiomic features reflecting TNBC subtypes. RESULTS: The components of the TIME score were established by m-IHC and demonstrated positive prognostic value for TNBC (p = 0.0047, 0.039, <0.0001 for DMFS, RFS, and OS). The score was calculated from several indicators, including Treg% in the tumor core (TC) or stromal area (SA), PD-L1+ cell% in the SA, CD3 + cell% in the TC, and PD-L1+ /CD8+ cells in the invasive margin and SA. According to the TNBC subtyping system, a few TIME indicators were significantly different in different subtypes and significantly correlated with MRI radiomic features reflecting TNBC subtypes. CONCLUSION: We demonstrated that the m-IHC-based quantitative score and indicators related to the spatial distribution and interactions of cells in the TIME can aid in the accurate diagnosis of TNBC in terms of prognosis and classification.

Artificial intelligence in breast imaging: Current situation and clinical challenges.

Breast cancer ranks among the most prevalent malignant tumours and is the primary contributor to cancer-related deaths in women. Breast imaging is essential for screening, diagnosis, and therapeutic surveillance. With the increasing demand for precision medicine, the heterogeneous nature of breast cancer makes it necessary to deeply mine and rationally utilize the tremendous amount of breast imaging information. With the rapid advancement of computer science, artificial intelligence (AI) has been noted to have great advantages in processing and mining of image information. Therefore, a growing number of scholars have started to focus on and research the utility of AI in breast imaging. Here, an overview of breast imaging databases and recent advances in AI research are provided, the challenges and problems in this field are discussed, and then constructive advice is further provided for ongoing scientific developments from the perspective of the National Natural Science Foundation of China.

Radiogenomic-based multiomic analysis reveals imaging intratumor heterogeneity phenotypes and therapeutic targets.

Intratumor heterogeneity (ITH) profoundly affects therapeutic responses and clinical outcomes. However, the widespread methods for assessing ITH based on genomic sequencing or pathological slides, which rely on limited tissue samples, may lead to inaccuracies due to potential sampling biases. Using a newly established multicenter breast cancer radio-multiomic dataset (n = 1474) encompassing radiomic features extracted from dynamic contrast-enhanced magnetic resonance images, we formulated a noninvasive radiomics methodology to effectively investigate ITH. Imaging ITH (IITH) was associated with genomic and pathological ITH, predicting poor prognosis independently in breast cancer. Through multiomic analysis, we identified activated oncogenic pathways and metabolic dysregulation in high-IITH tumors. Integrated metabolomic and transcriptomic analyses highlighted ferroptosis as a vulnerability and potential therapeutic target of high-IITH tumors. Collectively, this work emphasizes the superiority of radiomics in capturing ITH. Furthermore, we provide insights into the biological basis of IITH and propose therapeutic targets for breast cancers with elevated IITH.

Fully automating LI-RADS on MRI with deep learning-guided lesion segmentation, feature characterization, and score inference.

Introduction: Leveraging deep learning in the radiology community has great potential and practical significance. To explore the potential of fitting deep learning methods into the current Liver Imaging Reporting and Data System (LI-RADS) system, this paper provides a complete and fully automatic deep learning solution for the LI-RADS system and investigates its model performance in liver lesion segmentation and classification. Methods: To achieve this, a deep learning study design process is formulated, including clinical problem formulation, corresponding deep learning task identification, data acquisition, data preprocessing, and algorithm validation. On top of segmentation, a UNet++-based segmentation approach with supervised learning was performed by using 33,078 raw images obtained from 111 patients, which are collected from 2010 to 2017. The key innovation is that the proposed framework introduces one more step called feature characterization before LI-RADS score classification in comparison to prior work. In this step, a feature characterization network with multi-task learning and joint training strategy was proposed, followed by an inference module to generate the final LI-RADS score. Results: Both liver segmentation and feature characterization models were evaluated, and comprehensive statistical analysis was conducted with detailed discussions. Median DICE of liver lesion segmentation was able to achieve 0.879. Based on different thresholds, recall changes within a range of 0.7 to 0.9, and precision always stays high greater than 0.9. Segmentation model performance was also evaluated on the patient level and lesion level, and the evaluation results of (precision, recall) on the patient level were much better at approximately (1, 0.9). Lesion classification was evaluated to have an overall accuracy of 76%, and most mis-classification cases happen in the neighboring categories, which is reasonable since it is naturally difficult to distinguish LI-RADS 4 from LI-RADS 5. Discussion: In addition to investigating the performance of the proposed model itself, extensive comparison experiment was also conducted. This study shows that our proposed framework with feature characterization greatly improves the diagnostic performance which also validates the effectiveness of the added feature characterization step. Since this step could output the feature characterization results instead of simply generating a final score, it is able to unbox the black-box for the proposed algorithm thus improves the explainability.

The value of imaging combined with clinicopathological features in the diagnosis of high-risk breast lesions.

Background: The upgrade of high-risk breast lesions (HRLs) is closely related to subsequent treatment, but the current predictors for upgrade are limited to intratumoral features of single imaging mode. Methods: We retrospectively reviewed 230 HRLs detected by mammography, ultrasound, and magnetic resonance imaging (MRI) before biopsy at the Fudan University Cancer Hospital from January 2017 to March 2018. The clinical features, imaging data according to the Breast Imaging Reporting and Data System (BI-RADS) lexicon, and tumor upgrade situation were received. Based on the different risks of upgrade reported, the lesions were classified into high-risk I [HR-I, with atypical hyperplasia (AH)] and high-risk II (HR-II, without AH). We analyzed the association between clinicopathological and imaging factors and upgrade. We used the receiver operating characteristic (ROC) curve to compare the efficacy of three imaging modes for predicting upgrade. Results: We included 230 HRLs in 230 women in the study, and the overall upgrade rate was 20.4% (47/230). The upgrade rate was higher in HR-I compared to HR-II (38.5% vs. 4.1%, P<0.01). In patients with AH, estrogen receptor-positive (ER+) patients accounted for 81.0% (64/79). For all HRLs and HR-I, in clinical characteristics, age, maximum size of lesion, and menopausal status were significantly associated with upgrade (P<0.05). In imaging factors, MRI background parenchymal enhancement (BPE), signs of MRI and ultrasound were significantly correlated with upgrade (P<0.05). Patients with negative MRI or ultrasound manifestations had lower upgrade rates (P<0.01). For HR-II, only BPE showed a significant difference between groups (P=0.001). Multifactorial analysis of all HRLs showed that age and BPE were independent predictors of upgrade (P<0.01). The areas under the ROC cure (AUCs) for predicting upgrade in mammography, ultrasound, and MRI were 0.606, 0.590, and 0.913, respectively, indicating that MRI diagnosis was significantly better than mammography and ultrasound (P<0.001). Conclusions: HRLs with AH had a higher rate of upgrade and increased ER expression. Among three imaging modes, MRI was more effective than ultrasound and mammography in diagnosing the upgrade of HRLs. Older age and moderate to marked BPE can indicate malignant upgrade. MRI can provide a certain value for the diagnosis and follow-up of HRLs.

Postnatal expression of the lysine methyltransferase SETD1B is essential for learning and the regulation of neuron-enriched genes.

In mammals, histone 3 lysine 4 methylation (H3K4me) is mediated by six different lysine methyltransferases. Among these enzymes, SETD1B (SET domain containing 1b) has been linked to syndromic intellectual disability in human subjects, but its role in the mammalian postnatal brain has not been studied yet. Here, we employ mice deficient for Setd1b in excitatory neurons of the postnatal forebrain, and combine neuron-specific ChIP-seq and RNA-seq approaches to elucidate its role in neuronal gene expression. We observe that Setd1b controls the expression of a set of genes with a broad H3K4me3 peak at their promoters, enriched for neuron-specific genes linked to learning and memory function. Comparative analyses in mice with conditional deletion of Kmt2a and Kmt2b histone methyltransferases show that SETD1B plays a more pronounced and potent role in regulating such genes. Moreover, postnatal loss of Setd1b leads to severe learning impairment, suggesting that SETD1B-dependent regulation of H3K4me levels in postnatal neurons is critical for cognitive function.

Icariside II attenuates lipopolysaccharide-induced neuroinflammation through inhibiting TLR4/MyD88/NF-κB pathway in rats.

Inflammation in central nervous system (CNS) plays a vital role in neurodegenerative diseases such as Alzheimer's disease (AD), Parkinson's disease (PD), Lewy body dementia (DLB), HIV-related dementia and traumatic brain injury. Icariside II (ICS II), an active flavonoid compound derived from a Chinese herbal medicine Epimedium brevicornum Maxim, has been shown to possess a neuroprotective effect on AD model. However, whether ICS II has a directly protective effect on acute neuroinflammation remains still unclear. Therefore, the current study was designed to investigate the possible protective effect of ICS II on acute neuroinflammation induced by intracerebroventricular (ICV) injection of lipopolysaccharide (LPS), and further to explore its possible mechanism. After ICS II was prophylactically administered for 7 days before LPS injection, the rats were randomly divided into five groups as follows: sham group (n = 9), sham + ICS II-H (10 mg/kg) (n = 9), LPS (n = 14), LPS + ICS II-L (3 mg/kg) (n = 14), LPS + ICS II-H (10 mg/kg) (n = 14) groups, respectively. As expected, LPS injection exhibited neuronal morphological damage, and ionized calcium binding adapter molecule 1 (IBA-1) of microglia and glial fibrillary acidic protein (GFAP) of astrocyte were activated. However, pre-treatment with ICS II not only inhibited the activation of microglia and astrocyte, but also significantly reversed the expressions of inflammatory factors such as interleukin-1ß (IL-1ß), tumor necrosis factor (TNF-α), cyclooxygenase-2 (COX-2), as well as the expressions of Toll-Like receptor 4 (TLR4), myeloid differentiation factor 88 (MyD88) and TNF receptor associated factor 6 (TRAF6). Furthermore, ICS II inhibited the degradation of IκB and the following activation of NF-κB. Hence it is concluded that ICS II attenuates LPS-induced neuroinflammation through inhibiting TLR4/MyD88/NF-κB pathway in rats, and it has potential value as a new therapeutic agent to treat neuroinflammation-related diseases, such as AD.

Involvement of growth factors in diabetes mellitus and its complications: A general review.

Diabetes mellitus (DM) is a major endocrine metabolic disease and is marked by a lack of insulin. The complication of DM is one of the most difficult problems in medicine. The initial translational studies revealed that growth factors have a major role in integrating tissue physiology and in embryology as well as in growth, maturation and tissue repair. In some tissues affected by diabetes, growth factors are induced by a relative deficit or excess. Fibroblast growth factor 21 (FGF21) is a promising regulator of glucose and lipid metabolism with multiple beneficial effects including hypoglycemic and lipid-lowering. Vascular endothelial growth factor (VEGF) is a potent angiogenic and vascular permeability factor and is implicated in both of these complications in diabetes. Increase or decrease in the production of transforming growth factor-ß1 (TGF-ß1) has been associated with diabetic nephropathy and retinopathy. The insulin-like growth factor-I (IGF-I) is a naturally-occurring single chain polypeptide which has been widely used in the treatment of diabetic glomerular and renal tubular injuries. This review summarizes the recent evidences for an involvement of growth factors in diabetic complications, focusing on their emergence in sequence of events leading to vascular complications or their potential therapeutic role in these diseases. Growth factor therapy in diabetic foot ulcers is already a clinical reality. As methods to finely regulate growth factors in a tissue and time-specific manner are further developed and tested, regulation of the growth factor to normal level in vivo may well become a therapy to prevent and treat diabetic complications.

Icariside II, a Broad-Spectrum Anti-cancer Agent, Reverses Beta-Amyloid-Induced Cognitive Impairment through Reducing Inflammation and Apoptosis in Rats.

Beta-amyloid (Aß) deposition, associated neuronal apoptosis and neuroinflammation are considered as the important factors which lead to cognitive deficits in Alzheimer's disease (AD). Icariside II (ICS II), an active flavonoid compound derived from Epimedium brevicornum Maxim, has been extensively used to treat erectile dysfunction, osteoporosis and dementia in traditional Chinese medicine. Recently, ICS II attracts great interest due to its broad-spectrum anti-cancer property. ICS II shows an anti-inflammatory potential both in cancer treatment and cerebral ischemia-reperfusion. It is not yet clear whether the anti-inflammatory effect of ICS II could delay progression of AD. Therefore, the current study aimed to investigate the effects of ICS II on the behavioral deficits, Aß levels, neuroinflammatory responses and apoptosis in Aß25-35-treated rats. We found that bilateral hippocampal injection of Aß25-35 induced cognitive impairment, neuronal damage, along with increase of Aß, inflammation and apoptosis in hippocampus of rats. However, treatment with ICS II 20 mg/kg could improve the cognitive deficits, ameliorate neuronal death, and reduce the levels of Aß in the hippocampus. Furthermore, ICS II could suppress microglial and astrocytic activation, inhibit expression of IL-1ß, TNF-α, COX-2, and iNOS mRNA and protein, and attenuate the Aß induced Bax/Bcl-2 ratio elevation and caspase-3 activation. In conclusion, these results showed that ICS II could reverse Aß-induced cognitive deficits, possibly via the inhibition of neuroinflammation and apoptosis, which suggested a potential protective effect of ICS II on AD.

A virtual surgical training system that simulates cutting of soft tissue using a modified pre-computed elastic model.

This work presents a surgical training system that incorporates cutting operation of soft tissue simulated based on a modified pre-computed linear elastic model in the Simulation Open Framework Architecture (SOFA) environment. A precomputed linear elastic model used for the simulation of soft tissue deformation involves computing the compliance matrix a priori based on the topological information of the mesh. While this process may require a few minutes to several hours, based on the number of vertices in the mesh, it needs only to be computed once and allows real-time computation of the subsequent soft tissue deformation. However, as the compliance matrix is based on the initial topology of the mesh, it does not allow any topological changes during simulation, such as cutting or tearing of the mesh. This work proposes a way to modify the pre-computed data by correcting the topological connectivity in the compliance matrix, without re-computing the compliance matrix which is computationally expensive.

Random feature subspace ensemble based Extreme Learning Machine for liver tumor detection and segmentation.

This paper presents a new approach to detect and segment liver tumors. The detection and segmentation of liver tumors can be formulized as novelty detection or two-class classification problem. Each voxel is characterized by a rich feature vector, and a classifier using random feature subspace ensemble is trained to classify the voxels. Since Extreme Learning Machine (ELM) has advantages of very fast learning speed and good generalization ability, it is chosen to be the base classifier in the ensemble. Besides, majority voting is incorporated for fusion of classification results from the ensemble of base classifiers. In order to further increase testing accuracy, ELM autoencoder is implemented as a pre-training step. In automatic liver tumor detection, ELM is trained as a one-class classifier with only healthy liver samples, and the performance is compared with two-class ELM. In liver tumor segmentation, a semi-automatic approach is adopted by selecting samples in 3D space to train the classifier. The proposed method is tested and evaluated on a group of patients' CT data and experiment show promising results.

Prostate boundary segment extraction using cascaded shape regression and optimal surface detection.

In this paper, we proposed a new method (CSR+OSD) for the extraction of irregular open prostate boundaries in noisy extracorporeal ultrasound image. First, cascaded shape regression (CSR) is used to locate the position of prostate boundary in the images. In CSR, a sequence of random fern predictors are trained in a boosted regression manner, using shape-indexed features to achieve invariance against position variations of prostate boundaries. Afterwards, we adopt optimal surface detection (OSD) to refine the prostate boundary segments across 3D sections globally and efficiently. The proposed method is tested on 162 ECUS images acquired from 8 patients with benign prostate hyperplasia. The method yields a Root Mean Square Distance of 2.11±1.72 mm and a Mean Absolute Distance of 1.61±1.26 mm, which are lower than those of JFilament, an open active contour algorithm and Chan-Vese region based level set model, respectively.

Planning of hepatectomy for tumor resection: resection surface construction and optimization.

An interactive liver surgery planning system has been developed to construct and optimize the resection plan. With this system, the segmentation results of the liver and its components (such as tumors and vessels) are comprehensively visualized for surgeons to have an intuitive understanding of the internal anatomical structure of the liver. This system will also allow surgeons to interactively create and modify a resection plan on the virtual liver model. The resection surface, whose boundary is a closed curve, will be automatically constructed with the safe resection margins of tumors. Different from other systems, our developed system is able to generate the safety margins to all tumors. During surgery, a larger resection surface may cause potentially more bleeding and other complications. Therefore, area minimization is applied during the resection surface construction by adopting the minimal area mesh, which is a smooth surface with minimal area. After these virtual modifications, the resultant resection surface indicates the route to cut the liver for tumor removal. The volumes for both resected liver and residual liver are calculated for clinical decision making.

Segmentation of hepatic tumor from abdominal CT data using an improved support vector machine framework.

An improved support vector machine (SVM) framework has been developed to segment hepatic tumor from CT data. By this method, the one-class SVM (OSVM) and two-class SVM (TSVM) are connected seamlessly by a boosting tool, to tackle the tumor segmentation via both offline and online learning. An initial tumor region was first pre-segmented by an OSVM classifier. Then the boosting tool was employed to automatically generate the negative (non-tumor) samples, according to certain criteria. The pre-segmented initial tumor region and the non-tumor samples generated were used to train a TSVM) classifier. By the trained TSVM classifier, the final tumor lesion was segmented out. Tested on 16 sets of CT abdominal scans, quantitative results suggested that the developed method achieved significantly higher segmentation accuracy than the OSVM and TSVM classifiers.

Liver tumor detection and segmentation using kernel-based Extreme Learning Machine.

This paper presents an approach to detection and segmentation of liver tumors in 3D computed tomography (CT) images. The automatic detection of tumor can be formulized as novelty detection or two-class classification issue. The method can also be used for tumor segmentation, where each voxel is to be assigned with a correct label, either a tumor class or nontumor class. A voxel is represented with a rich feature vector that distinguishes itself from voxels in different classes. A fast learning algorithm Extreme Learning Machine (ELM) is trained as a voxel classifier. In automatic liver tumor detection, we propose and show that ELM can be trained as a one-class classifier with only healthy liver samples in training. It results in a method of tumor detection based on novelty detection. We compare it with two-class ELM. To extract the boundary of a tumor, we adopt the semi-automatic approach by randomly selecting samples in 3D space within a limited region of interest (ROI) for classifier training. Our approach is validated on a group of patients' CT data and the experiment shows good detection and encouraging segmentation results.

Indication of lower neck irradiation in nasopharyngeal carcinoma without nodal metastasis: the potential impact of tumor volume.

BACKGROUND: Elective radiation of lower neck is controversial for nasopharyngeal carcinoma (NPC) without lymph node metastasis (N0 disease). Tumor volume is an important prognostic indicator. The objective of this study is to explore the potential impact of tumor volume on the indication of the lower neck irradiation for N0-NPC, by a qualitative evaluation of the relationship between tumor volume and nodal metastasis. METHODS: Magnetic resonance (MR) images of 99 consecutive patients with NPC who underwent treatment were retrospectively reviewed. Primary tumor volumes of NPC were semi-automatically measured, nodal metastases were N-classified and neck level involvements were examined. Distributions of tumor volumes among N-category-based groups and distributions of N-categories among tumor volume-based groups were analyzed, respectively. RESULTS: The numbers of patients with N0 to N3 disease were 12, 39, 32, and 16, respectively. The volumes of primary tumor were from 3.3 to 89.6 ml, with a median of 17.1 ml. For patients with nodal metastasis, tumor volume did not increase significantly with the advancing of N-category (P > 0.05). No significant difference was found for the distribution of N1, N2, and N3 categories among tumor volume-based groups (P > 0.05). Nevertheless patients with nodal metastasis had significantly larger tumor volumes than those without metastasis (P < 0.05). Patients with larger tumor volumes were associated with an increased incidence of nodal metastasis. CONCLUSIONS: Certain positive correlations existed between tumor volume and the presence of nodal metastasis. The tumor volume (>10 ml) is a potential indicator for the lower neck irradiation for N0-NPC.