Deception detection plays a crucial role in criminal investigation.Videos contain a wealth of information regarding apparent and physiological changes in individuals,and thus can serve as an effective means of decepti...Deception detection plays a crucial role in criminal investigation.Videos contain a wealth of information regarding apparent and physiological changes in individuals,and thus can serve as an effective means of deception detection.In this paper,we investigate video-based deception detection considering both apparent visual features such as eye gaze,head pose and facial action unit(AU),and non-contact heart rate detected by remote photoplethysmography(rPPG)technique.Multiple wrapper-based feature selection methods combined with the K-nearest neighbor(KNN)and support vector machine(SVM)classifiers are employed to screen the most effective features for deception detection.We evaluate the performance of the proposed method on both a self-collected physiological-assisted visual deception detection(PV3D)dataset and a public bag-oflies(BOL)dataset.Experimental results demonstrate that the SVM classifier with symbiotic organisms search(SOS)feature selection yields the best overall performance,with an area under the curve(AUC)of 83.27%and accuracy(ACC)of 83.33%for PV3D,and an AUC of 71.18%and ACC of 70.33%for BOL.This demonstrates the stability and effectiveness of the proposed method in video-based deception detection tasks.展开更多
Background： Rosai-Dorfman disease （RDD） is typically characterized by painless bilateral and symmetrical cervical lymphadenopathy, with associated fever and leukocytosis. The aim of the current study was to summari...Background： Rosai-Dorfman disease （RDD） is typically characterized by painless bilateral and symmetrical cervical lymphadenopathy, with associated fever and leukocytosis. The aim of the current study was to summarize the clinical features and imaging characteristics of RDD, in an effort to improve its diagnostic accuracy. Methods： The study was analyzed from 32 patients between January 2011 and December 2017; of these, 16 patients had pathologically diagnosed RDD, eight had pathologically diagnosed meningioma, and eight pathologically diagnosed lymphoma. All patients underwent computed tomography and magnetic resonance imaging （MRI）. Clinical features and imaging characteristics of RDD were analyzed retrospectively. The mean apparent diffusion coefficient （ADC） values of lesions at different sites were measured, and one-way analysis of variance and the least significant difference t-test were used to compare the differences between groups and draw receiver operating characteristic curves. The tumors were excised for biopsy and analyzed using immunohistochemistry. Results： The mean ADCs were （0.81 ± 0.10） × 10^-3mm^2/s for intercranial RDD, （0.73 ± 0.05） × 10^-3mm^2/s for nasopharyngeal RDD, （0.74 ± 0.11） × 10^-3mm2/s for bone RDD, and （0.71 ± 0.04） × 10^-3mm2/s for soft-tissue RDD. The optimum ADC to distinguish intracranial RDD from lymphoma was 0.79 × 10^-3mm2/s （62.5% sensitivity and 100% specificity） and to distinguish meningioma from intracranial RDD was 0.92 × 10^-3mm^2/s （62.5% sensitivity and 100% specificity）. Levels of C-reactive protein, erythrocyte sediment rate and D-dimer were significantly elevated （81%, 87%, and 75%, respectively）. On immunohistochemistry, RDD was positive for both S-100 and CD68 proteins but negative for CD1a. Conclusions： Conventional MRI, combined with diffusion-weighted imaging and ADC mapping, is an important diagnostic tool in evaluating RDD patients. An accurate diagnosis of RDD should consider the clinical features,展开更多
基金National Natural Science Foundation of China(No.62271186)Anhui Key Project of Research and Development Plan(No.202104d07020005)。
文摘Deception detection plays a crucial role in criminal investigation.Videos contain a wealth of information regarding apparent and physiological changes in individuals,and thus can serve as an effective means of deception detection.In this paper,we investigate video-based deception detection considering both apparent visual features such as eye gaze,head pose and facial action unit(AU),and non-contact heart rate detected by remote photoplethysmography(rPPG)technique.Multiple wrapper-based feature selection methods combined with the K-nearest neighbor(KNN)and support vector machine(SVM)classifiers are employed to screen the most effective features for deception detection.We evaluate the performance of the proposed method on both a self-collected physiological-assisted visual deception detection(PV3D)dataset and a public bag-oflies(BOL)dataset.Experimental results demonstrate that the SVM classifier with symbiotic organisms search(SOS)feature selection yields the best overall performance,with an area under the curve(AUC)of 83.27%and accuracy(ACC)of 83.33%for PV3D,and an AUC of 71.18%and ACC of 70.33%for BOL.This demonstrates the stability and effectiveness of the proposed method in video-based deception detection tasks.
文摘Background： Rosai-Dorfman disease （RDD） is typically characterized by painless bilateral and symmetrical cervical lymphadenopathy, with associated fever and leukocytosis. The aim of the current study was to summarize the clinical features and imaging characteristics of RDD, in an effort to improve its diagnostic accuracy. Methods： The study was analyzed from 32 patients between January 2011 and December 2017; of these, 16 patients had pathologically diagnosed RDD, eight had pathologically diagnosed meningioma, and eight pathologically diagnosed lymphoma. All patients underwent computed tomography and magnetic resonance imaging （MRI）. Clinical features and imaging characteristics of RDD were analyzed retrospectively. The mean apparent diffusion coefficient （ADC） values of lesions at different sites were measured, and one-way analysis of variance and the least significant difference t-test were used to compare the differences between groups and draw receiver operating characteristic curves. The tumors were excised for biopsy and analyzed using immunohistochemistry. Results： The mean ADCs were （0.81 ± 0.10） × 10^-3mm^2/s for intercranial RDD, （0.73 ± 0.05） × 10^-3mm^2/s for nasopharyngeal RDD, （0.74 ± 0.11） × 10^-3mm2/s for bone RDD, and （0.71 ± 0.04） × 10^-3mm2/s for soft-tissue RDD. The optimum ADC to distinguish intracranial RDD from lymphoma was 0.79 × 10^-3mm2/s （62.5% sensitivity and 100% specificity） and to distinguish meningioma from intracranial RDD was 0.92 × 10^-3mm^2/s （62.5% sensitivity and 100% specificity）. Levels of C-reactive protein, erythrocyte sediment rate and D-dimer were significantly elevated （81%, 87%, and 75%, respectively）. On immunohistochemistry, RDD was positive for both S-100 and CD68 proteins but negative for CD1a. Conclusions： Conventional MRI, combined with diffusion-weighted imaging and ADC mapping, is an important diagnostic tool in evaluating RDD patients. An accurate diagnosis of RDD should consider the clinical features,
