Search

Abstract Background Despite the utilization of immune checkpoint inhibitors (ICIs) in treating numerous types of cancers being approved, their efficacy in tumor control in the clinic is not satisfactory. Since adoptive cell therapy (ACT) can alter the tumor microenvironment, we hypothesized that ACT potentially synergized with ICI in tumor control and examined this hypothesis via a murine allograft model. Methods Female C57BL/6 mice were stimulated with interleukin 15 and granulocyte monocyte-colony stimulating factor, followed by collecting their bone marrow cells for murine NKDC cultivation. Then, female C57BL/6 mice, inoculated with lymphoma cancer cell line E.G7-OVA, were administrated with murine NKDC cells, murine anti-program cell death ligand-1 antibody (α-mPD-L1), or both for 28 days. After 28 days of treatment, mice were sacrificed whose inoculated tumors, spleen, sentinel lymph nodes, and peripheral blood were collected to measure tumor size, lymphocyte infiltration, and change of immune cell profile. Results Combined treatment of NKDCs with α-mPD-L1 exhibited significantly stronger tumor control efficacy than treatment of NKDCs or α-mPD-L1 alone. NKDCs/α-mPD-L1 combination increased migration of dendritic cells, CD4, CD8 T cells, and activated CD8 T cells to the tumor-bedding site, and promoted endogenous tumor-specific cytotoxic T-cell response. Conclusion The current study confirmed our hypothesis that combining NKDC ACT with ICI therapy can potentiate tumor control efficacy by manipulating the tumor microenvironment. This study provided a novel circumstance on tumor immunotherapy.

Camera, inertial measurement unit (IMU), and ultra-wideband (UWB) sensors are commonplace solutions to unmanned aerial vehicle (UAV) localization problems. The performance of a localization system can be improved by integrating observations from different sensors. In this paper, we propose a learning-based UAV localization method using the fusion of vision, IMU, and UWB sensors. Our model consists of visual–inertial (VI) and UWB branches. We combine the estimation results of both branches to predict global poses. To evaluate our method, we augment a public VI dataset with UWB simulations and conduct a real-world experiment. The experimental results show that our method provides more robust and accurate results than VI/UWB-only localization. Our codes and data are available at https://imlabntu.github.io/VIUNet/.

Abstract Background The prognosis of unfavorable cancer of unknown primary is extremely poor. This is the first report to compared the treatment results between generations of CUP and examined prognostic factors. Methods This retrospective single‐center cohort study enrolled 68 patients with newly diagnosed unfavorable cancer of unknown primary at Taipei Veteran General Hospital from 2017 to 2020 as study cohort and 167 patients from 2000 to 2009 as historical cohort. Results The median overall survival was 4.3 months in the study cohort (95% CI, 2.7–6.2 months) and 4.5 months in the historical cohort (95% CI, 3.0–5.5 months; p = 0.858). Eleven patients in the study cohort received immunotherapy. The disease control rates were 45%. Multivariate analysis showed that an Eastern Cooperative Oncology Group score > 1 and a C‐reactive protein level > 1 correlated with poor survival. A new prognostic stratification model was constructed by using Eastern Cooperative Oncology Group score and C‐reactive protein values. The good‐, intermediate‐, and poor‐risk groups had distinct median overall survival of 18.3, 7.0 and 1.2 months, respectively (area under the curve, 0.817; p < 0.001). Conclusion The outcome of unfavorable cancer of unknown primary has not changed much over the last 20 years. The application of a new prognostic stratification model can further stratify unfavorable cancer of unknown primary.

Abstract Background Vascular endothelial growth factor (VEGF) plays a role in the tumor microenvironment. Sorafenib, which inhibits the VEGF pathway, has an immune-modulation function but lacks substantial clinical data. This study aims to explore the efficacy of anti-PD-1 combined sorafenib in advanced hepatocellular carcinoma (HCC). Methods HCC patients who underwent anti-PD-1 treatment at Taipei Veterans General Hospital (Taipei, Taiwan) between January 2016 and February 2019 were reviewed. The efficacy was compared between groups after propensity-score matching. Results There were 173 HCC patients receiving anti-PD-1. After excluding unsuitable cases, 140 patients were analyzed, of which 58 received combination therapy and 82 received anti-PD-1 alone. The combination therapy had a trend of higher CR rate (8.6% vs. 4.9%, ns.), ORR (22.4% vs. 19.5%, ns.) and significantly higher DCR (69.0% vs. 37.8%, p 10, > 50 and > 66%), the higher the ORR (70, 80 and 92%) and CR rates (30, 35 and 58%) were achieved at day 28. Conclusions This is the first study to demonstrate the combination of anti-PD-1 and sorafenib had better efficacy and survival benefit. A prospective randomized study is needed to confirm this finding.

Purpose: Immune checkpoint inhibitors (ICIs) alone or in combination with chemotherapy can improve the limited efficacy of colorectal cancer (CRC) immunotherapy. CX-5461 causes substantial DNA damage and genomic instability and can increase ICIs’ therapeutic efficacies through tumor microenvironment alteration. Results: We analyzed whether CX-5461 enhances ICIs’ effects in CRC and discovered that CX-5461 causes severe DNA damage, including cytosolic dsDNA appearance, in various human and mouse CRC cells. Our bioinformatics analysis predicted CX-5461-based interferon (IFN) signaling pathway activation in these cells, which was verified by the finding that CX-5461 induces IFN-α and IFN-β secretion in these cells. Next, cGAMP, phospho-IRF3, CCL5, and CXCL10 levels exhibited significant posttreatment increases in CRC cells, indicating that CX-5461 activates the cGAS-STING-IFN pathway. CX-5461 also enhanced PD-L1 expression through STAT1 activation. CX-5461 alone inhibited tumor growth and prolonged survival in mice. CX-5461+anti-PD-1 or anti-PD-L1 alone exhibited synergistic growth-suppressive effects against CRC and breast cancer. CX-5461 alone or CX-5461+anti-PD-1 increased cytotoxic T-cell numbers and reduced myeloid-derived suppressor cell numbers in mouse spleens. Conclusions: Therefore, clinically, CX-5461 combined with ICIs for CRC therapy warrants consideration because CX-5461 can turn cold tumors into hot ones.

Immune checkpoint inhibitors (ICI) have been applied in treating advanced hepatocellular carcinoma (aHCC) patients, but few patients exhibit stable and lasting responses. Moreover, identifying aHCC patients suitable for ICI treatment is still challenged. This study aimed to evaluate whether dissecting peripheral immune cell subsets by Mann-Whitney U test and artificial intelligence (AI) algorithms could serve as predictive biomarkers of nivolumab treatment for aHCC. Disease control group carried significantly increased percentages of PD-L1+ monocytes, PD-L1+ CD8 T cells, PD-L1+ CD8 NKT cells, and decreased percentages of PD-L1+ CD8 NKT cells via Mann-Whitney U test. By recursive feature elimination method, five featured subsets (CD4 NKTreg, PD-1+ CD8 T cells, PD-1+ CD8 NKT cells, PD-L1+ CD8 T cells and PD-L1+ monocytes) were selected for AI training. The featured subsets were highly overlapping with ones identified via Mann-Whitney U test. Trained AI algorithms committed valuable AUC from 0.8417 to 0.875 to significantly separate disease control group from disease progression group, and SHAP value ranking also revealed PD-L1+ monocytes and PD-L1+ CD8 T cells exclusively and significantly contributed to this discrimination. In summary, the current study demonstrated that integrally analyzing immune cell profiling with AI algorithms could serve as predictive biomarkers of ICI treatment.

Background and Aims: Current prediction models for early recurrence of hepatocellular carcinoma (HCC) after surgical resection remain unsatisfactory. The aim of this study was to develop evolutionary learning-derived prediction models with interpretability using both clinical and radiomic features to predict early recurrence of HCC after surgical resection. Methods: Consecutive 517 HCC patients receiving surgical resection with available contrast-enhanced computed tomography (CECT) images before resection were retrospectively enrolled. Patients were randomly assigned to a training set (n = 362) and a test set (n = 155) in a ratio of 7:3. Tumor segmentation of all CECT images including noncontrast phase, arterial phase, and portal venous phase was manually performed for radiomic feature extraction. A novel evolutionary learning-derived method called genetic algorithm for predicting recurrence after surgery of liver cancer (GARSL) was proposed to design prediction models for early recurrence of HCC within 2 years after surgery. Results: A total of 143 features, including 26 preoperative clinical features, 5 postoperative pathological features, and 112 radiomic features were used to develop GARSL preoperative and postoperative models. The area under the receiver operating characteristic curves (AUCs) for early recurrence of HCC within 2 years were 0.781 and 0.767, respectively, in the training set, and 0.739 and 0.741, respectively, in the test set. The accuracy of GARSL models derived from the evolutionary learning method was significantly better than models derived from other well-known machine learning methods or the early recurrence after surgery for liver tumor (ERASL) preoperative (AUC = 0.687, p < 0.001 vs. GARSL preoperative) and ERASL postoperative (AUC = 0.688, p < 0.001 vs. GARSL postoperative) models using clinical features only. Conclusion: The GARSL models using both clinical and radiomic features significantly improved the accuracy to predict early recurrence of HCC after surgical resection, which was significantly better than other well-known machine learning-derived models and currently available clinical models.

Background/Purpose: Recent progress in cancer immunology provides more insight in immune evasion of cancer cells. Cancer cells may achieve immune evasion through several ways including ineffective antigen presentation, T cell checkpoint utilization, immunosuppressive cytokines secretion and immunosuppressive cells recruitment. However, few literatures mentioned about the change of peripheral blood immune cells in advanced hepatocellular carcinoma (HCC) patients. To answer this question, we initiated a pilot study through detailed flow cytometry. Methods: We enrolled patients with advanced HCC patients who had informed consent to the collection of their peripheral blood. We also recruited healthy individuals for the control group. Using flow cytometry, we analyzed lymphocyte subclasses and the PD-1 or PD-L1 positivity of immune cells in peripheral blood from HCC patients and healthy individuals. Results: Twenty-four HCC patients were enrolled and twenty healthy individuals were enrolled. Most of the HCC patients were HBV carrier (58.3%), and the mean age was 61 years old. Among 55 immune cell parameters we examined in peripheral blood, 16 were significantly different between advanced HCC patients and healthy individuals by univariate analysis. Multivariate analysis was then conducted by fitting logistic regression model and showed that CD69−CD25− Naïve CD4αβT cell percentage and dendritic cell percentage can reasonably predict the advanced HCC status from peripheral blood. By our regression model, the adjusted generalized R2 = 0.918 and the estimated area under the Receiver Operating Characteristic (ROC) curve was 0.99. Conclusion: CD69−CD25− Naïve CD4αβT cell percentage and dendritic cell percentage in peripheral blood are highly correlated with the advanced HCC status. The change may result from immune evasion initiated by hepatocellular carcinoma cells and further investigation is warranted. Validation study is ongoing and this mechanism may be utilized to treat advanced HCC patient in the future.

Background: The response rate to sorafenib is limited for unresectable hepatocellular carcinoma (HCC). Little is known about the long-term outcomes of objective responders. The role of second-line therapies on the survival of sorafenib-responders is unclear. We aimed to delineate the long-term outcomes and the role of subsequent treatment after responding to sorafenib. Methods: From September 2012 to December 2019, 922 patients who received sorafenib treatment for unresectable HCC were retrospectively reviewed. Of these, 21 (2.3%) achieved a complete response (CR) and 54 (5.9%) had a partial response (PR) based on mRECIST criteria. Factors associated with survivals were analyzed. Results: During the median follow-up of 35.3 months, the median duration of response was 18.3 months (range: 2.3–45.5) for patients achieving CR and 10.0 months (range: 1.9–60.3) for PR. The median overall survival (OS) was 39.5 months [95% confidence interval (CI): 28.4–50.5] including values not yet estimable for CR and 25.8 months for PR. Patients who experienced treatment-related adverse events (TRAEs) had better median OS than those without (44.9 versus 18.1 months, p = 0.003). Eventually, 53 patients developed tumor progression; 30 patients received second-line systemic treatment including nivolumab ( n = 8), regorafenib ( n = 15), and chemotherapy ( n = 7). Sorafenib–nivolumab sequential therapy provided the best median OS versus sorafenib–regorafenib and sorafenib–chemotherapy in these patients (55.8, 39.5, and 25.5 months), respectively. Conclusions: The response is durable for advanced HCC patients with CR or PR to sorafenib. Subsequent immunotherapy seems to provide the best survival. This information is important for characterizing outcomes of sorafenib-responders and the choice of sequential treatment.

Fourier-domain optical coherence tomography (FD-OCT), including spectral-domain OCT (SD-OCT) and swept-source OCT (SS-OCT), allows the volumetric imaging of the tissue architecture with a faster speed and higher detection sensitivity than does time-domain OCT. Although the hardware implementations of SD-OCT and SS-OCT are different, these technologies share very similar signal processing steps for image reconstruction. In this study, we developed hardware implementations and software architectures to design a generic framework for FD-OCT. For SD-OCT systems, an external synchronization approach was used to realize a data acquisition schematic similar to that used in SS-OCT by carefully managing the timing clocks in the detection unit and for the waveform generation. In addition, by utilizing modules and factory concepts, a software engine can be developed that supports various acquisition devices and software operations or image processing functions with high operation flexibility while maintaining its robustness. Data processing and data saving were optimized using the parallel computing method with the OpenMP library and by leveraging the parallelism within the acquired data, respectively.

Abstract Our previous study showed Tai‐Chi‐Chuan (TCC) training, a moderate exercise, at school improved pulmonary function and inflammation profiles in children with mild asthma. However, habitual practice is hard to maintain with the lack of continuous family and peer support. We investigated whether parental intervention with social media could enhance children's adherence to exercise at home and improve asthmatic outcome measures. Parents were opted to attend a 12‐week TCC classroom training, supervise home practice, and report to a four‐step web‐based social media platform to stay updated and motivated through logging activity and tracking competition. Fractional exhaled nitric oxide (FeNO), FEV1/FVC and peak expiratory flow (PEF) were measured before and after 12 weeks of training. Fifty‐three asthmatic children were allocated into non‐TCC (control, n = 12), TCC groups with moderate‐to‐severe (TCC‐S, n = 26) and mild‐to‐moderate (TCC‐M, n = 15) asthma. We found both TCC groups exhibited better pulmonary function than the non‐TCC control. TCC increased FVC in mild‐to‐moderate asthma children while more pronounced improvement in FEV1, FEV1/FVC, PEF and FeNO was noticed in moderate‐to‐severe asthmatic children. All TCC subjects retained greater participation and better interaction online except for low‐ranking families who dramatically dropped their practice 9 weeks later. For asthmatic children, moderate exercise improves pulmonary functions in a severity‐dependent fashion. Parent‐based Learn‐Practice‐Persuade‐Award wheel is a useful platform to motivate children engagement in physical activity. Classical social persuasive skills could enhance general parent‐child relationship but tend to decrease in persuasiveness over time in low‐ranking families.

Overall survival (OS) in patients with advanced hepatocellular carcinoma (HCC) has improved in the era of multi-line sequential therapy. The application of antiviral therapy and its impact on survival for patients with HBV-related HCC needs to be reassessed. The aim of this study was to evaluate the application and impact of antiviral therapy on survival for patients with HBV-related HCC receiving tyrosine kinase inhibitor (TKI) therapy. Patients with advanced HBV-related HCC treated with sorafenib or lenvatinib as first-line therapy with (n = 377) and without (n = 182) nucleos(t)ide analogue (NUC) therapy were retrospectively enrolled. Prognostic factors of OS were evaluated. Secular trends in the increased application of NUC therapy and improved survival were observed in the last decade. The HBV reactivation rate in patients without NUC therapy was 6.6%. By multivariate analysis, baseline low HBV viral load, achieving undetectable HBV DNA after TKI therapy, and ability to receive second-line therapy were found to be independent predictors of OS. In subgroup patients with NUC therapy, starting NUC before TKI was associated with a better OS. In conclusion, the application of antiviral therapy for patients with HBV-related HCC receiving TKI therapy has increased over time. Achieving complete virological suppression may contribute to a better OS in patients with advanced HBV-related HCC.

Absolute pose regression (APR) for camera localization is a single-shot approach that encodes the information of a 3D scene in an end-to-end neural network. The camera pose result of APR methods can be observed as the linear combination of the base poses. Previous APR methods’ base poses are learned from training data. However, the training data can limit the performance of the methods, which cannot be generalized to cover the entire scene. To solve this issue, we use handcrafted base poses instead of learning-based base poses, which prevents overfitting the camera poses of the training data. Moreover, we use a dual-stream network architecture to process color and depth images separately to get more accurate localization. On the 7 Scenes dataset, the proposed method is among the best in median rotation error, and in median translation error, it outperforms previous APR methods. On a more difficult dataset—Oxford RobotCar dataset, the proposed method achieves notable improvements in median translation and rotation errors compared to the state-of-the-art APR methods.

Abstract Objective Immune checkpoint inhibitors (ICIs) have become the standard of care in various cancers, although the predictive tool is still unknown. Methods This study aimed to develop a novel gene panel by selecting DNA damage response (DDR) genes from the Catalogue of Somatic Mutations in Cancer (COSMIC) databank and validating them in previously reported cohorts. This association between DDR gene mutations and tumor mutation burden or microsatellite status was analysed from The Cancer Genome Atlas (TCGA) databank. Furthermore, we made the gene panel clinically accessible and predicted the response in clinical patients receiving ICIs by using cell‐free DNA. Results The top 20 mutated DDR genes in various cancers (total 37 genes) were taken from the COSMIC databank, and the DDR genes found to individually predict a response rate > 50% in Van Allen's cohort were selected (Science, 350, 2015 and 207). Eighteen DDR genes were selected as the gene panel. The prevalence and predicted response rate were validated in the other three reported cohorts. Tumor mutational burden‐high was positively associated with mutations of the 18 DDR genes for most cancers. We used cell‐free DNA to test the DDR gene panel and validated by our patients receiving ICIs. This DDR gene panel accounted for approximately 30% of various cancers, achieving a predicted response rate of approximately 60% in patients with a mutated gene panel receiving ICIs. Conclusion This gene panel is a novel and reliable tool for predicting the response to ICIs in cancer patients and guides the appropriate administration of ICIs in clinical practice.

Sleep quality is known to have a considerable impact on human health. Recent research shows that head and body pose play a vital role in affecting sleep quality. This paper presents a deep multi-task learning network to perform head and upper-body detection and pose classification during sleep. The proposed system has two major advantages: first, it detects and predicts upper-body pose and head pose simultaneously during sleep, and second, it is a contact-free home security camera-based monitoring system that can work on remote subjects, as it uses images captured by a home security camera. In addition, a synopsis of sleep postures is provided for analysis and diagnosis of sleep patterns. Experimental results show that our multi-task model achieves an average of 92.5% accuracy on challenging datasets, yields the best performance compared to the other methods, and obtains 91.7% accuracy on the real-life overnight sleep data. The proposed system can be applied reliably to extensive public sleep data with various covering conditions and is robust to real-life overnight sleep data.

Cholangiocarcinoma is the most common primary malignant tumor of the bile duct. The current standard first-line treatment for advanced or metastatic cholangiocarcinoma is gemcitabine and cisplatin. However, few effective treatment choices exist for refractory cholangiocarcinoma, and additional therapeutic drugs are urgently required. Our previous work demonstrated that the ALDH isoform 1A3 plays a vital role in the malignant behavior of cholangiocarcinoma and may serve as a new therapeutic target. In this study, we found a positive correlation between ALDH1A3 protein expression levels and the cell migration abilities of three cholangiocarcinoma cell lines, which was verified using ALDH1A3-overexpressing and ALDH1A3-knockdown clones. We also used ALDH1A3-high and ALDH1A3-low populations of cholangiocarcinoma cell lines from the library of integrated network-based cellular signatures (LINCS) program and assessed the effects of ruxolitinib, a commercially available JAK2 inhibitor. Ruxolitinib had a higher cytotoxic effect when combined with gemcitabine. Furthermore, the nuclear translocation STAT1 and STAT3 heterodimers were markedly diminished by ruxolitinib treatment, possibly resulting in decreased ALDH1A3 activation. Notably, ruxolitinib alone or combined with gemcitabine led to significantly reduced tumor size and weight. Collectively, our studies suggest that ruxolitinib might suppress the ALDH1A3 activation through the JAK2/STAT1/3 pathway in cholangiocarcinoma, and trials should be undertaken to evaluate its efficacy in clinical therapy.

The classification of pathological tumor-node-metastasis (pTNM) staging of gastric cancer was revised in the 8th American Joint Committee on Cancer (AJCC) edition. The major revision was the separation of pN3a and pN3b in the pTNM staging. The current study evaluated the prognostic impact of this change.A total of 1,517 patients who underwent curative surgery for gastric cancer with a retrieved lymph node number ≥15 at our institution from January 1995 to December 2011 were enrolled. Survival was compared for the disease classified according to both the 7th and 8th editions.After separation of pN3a and pN3b in the pTNM stage definition, the 8th edition still provides significant survival differences between each stage. The multivariate analysis demonstrated that the pTNM stage in both the 7th and 8th editions was an independent prognostic factors of overall survival and disease-free survival. The 8th edition has a better homogeneity than the 7th edition with a significantly higher likelihood ratio chi-square test. Regarding the OS and DFS, the time-dependent receiver operating characteristic (ROC) curves of the two staging systems are almost overlapping, indicating that the prognostic performance is comparable between the two staging systems.Both the 7th and 8th edition-based stages are independent prognostic factors for gastric cancer. The 8th edition has a better homogeneity than the 7th edition; the 8th edition provides discriminant survival differences among each pTNM stage that are comparable to those in the 7th edition.

The treatment of multiple myeloma (MM) has made greatly advances over the last decade, nearly doubling the overall survival time after the use of novel agents such as thalidomide, lenalidomide and bortezomib. However, the improvement of overall survival has occasionally led to clinical scenarios that were rarely recognized before, for example, light chain escape and extramedullary plasmacytoma (EMP). Here we report an MM patient who relapsed and manifested simultaneously with both a light chain escape and a disseminated pattern of EMP after several different modes of treatment, including conventional chemotherapy, autologous peripheral blood stem cell (PBSC) transplantation, and novel agents. The result was disappointing in this case, which taught us several important things: First, intact immunoglobulin may not be a reliable marker in the surveillance of pretreated multiple myeloma. Various tests for renal function, skeletal condition, and serum free light chain should also be carried out for a comprehensive evaluation. Second, EMP remains a tremendous challenge to the clinician, especially when refractory to current novel agents. Further researches are required to clarify the optimal treatment of such a condition in order to help more patients.

Female breast cancer patients have an increased risk of developing subsequent malignant diseases, but this issue is rarely discussed in regards to male breast cancer patients. Thus, we conducted a national survey that included 100,915 female and 578 male breast cancer patients to investigate the risk of second primary malignancy (SPM). During a follow-up period that included 529,782 person-years, 3,153 cases of SPM developed. Compared with the general population, the standardized incidence ratio (SIR) of SPM in breast cancer patients was 1.51 [95% confidence interval (CI): 1.46-1.56]. The observed risk was significantly higher in male patients (SIR 2.17, 95% CI 1.70-2.73) and in patients whose age at breast cancer diagnosis was 40 years or younger (SIR 3.39, 95% CI 2.80-4.07), comparing to age-matched general population. Compared with the overall female population, the SIRs of female breast cancer patients with uterine (SIR: 2.66, 95% CI: 2.37-2.98), thyroid (SIR: 2.30, 95% CI: 2.02-2.62), and bone and soft tissue (SIR: 2.16, 95% CI: 1.56-2.91) cancers were significantly increased. Male breast cancer patients also displayed significantly higher SIRs for thyroid (SIR: 13.2, 95% CI: 1.60-47.69), skin (SIR: 8.24, 95% CI: 3.02-17.94) and head and neck (SIR: 4.41, 95% CI: 2.35-7.54) cancers. Among breast cancer patients, risk factors significantly associated with SPM included male gender, older age, chemotherapy treatment and comorbidity with liver cirrhosis. From our analysis, we concluded that the risk of SPM was significantly higher for both male and female breast cancer patients compared with the general population, suggesting that more intensive surveillance may be needed, especially in high-risk patients.

BACKGROUND & AIMS: To evaluate the risk of depressive disorders among non-alcoholic patients by using the Taiwan National Health Insurance Research Database (NHIRD). METHODS: We conducted a retrospective study of a matched cohort of 52 725 participants (10 545 non-alcoholic cirrhotic patients and 42 180 control patients) who were selected from the NHIRD. Patients were observed for a maximum of 11 years to determine the rates of newly onset depressive disorders, and Cox regression was used to identify the risk factors associated with depressive disorders in cirrhotic patients. RESULTS: During the 11-year follow-up period, 395 (3.75%) non-alcoholic cirrhotic patients and 1 183 (2.80%) control patients were diagnosed with depressive disorders. The incidence risk ratio of depressive disorders between non-alcoholic cirrhotic patients and control patients was 1.76 (95% CI, 1.57-1.98, P

Object movie refers to a set of images captured from different perspectives around a 3D object. Object movie provides a good representation of a physical object because it can provide 3D interactive viewing effect, but does not require 3D model reconstruction. In this paper, we propose an efficient approach for content-based object movie retrieval. In order to retrieve the desired object movie from the database, we first map an object movie into the sampling of a manifold in the feature space. Two different layers of feature descriptors, dense and condensed, are designed to sample the manifold for representing object movies. Based on these descriptors, we define the dissimilarity measure between the query and the target in the object movie database. The query we considered can be either an entire object movie or simply a subset of views. We further design a relevance feedback approach to improving retrieved results. Finally, some experimental results are presented to show the efficacy of our approach.

This paper proposes an approach based on a state-space model for learning the user concepts in image retrieval. We first design a scheme of region-based image representation based on concept units, which are integrated with different types of feature spaces and with different region scales of image segmentation. The design of the concept units aims at describing similar characteristics at a certain perspective among relevant images. We present the details of our proposed approach based on a state-space model for interactive image retrieval, including likelihood and transition models, and we also describe some experiments that show the efficacy of our proposed model. This work demonstrates the feasibility of using a state-space model to estimate the user intuition in image retrieval.