Your browser doesn't support javascript.
loading
Mostrar: 20 | 50 | 100
Resultados 1 - 10 de 10
Filtrar
1.
Eur Radiol ; 2022 Mar 08.
Artigo em Inglês | MEDLINE | ID: mdl-35258676

RESUMO

OBJECTIVES: To build an artificial intelligence (AI) system to classify benign and malignant non-mass enhancement (NME) lesions using maximum intensity projection (MIP) of early post-contrast subtracted breast MR images. METHODS: This retrospective study collected 965 pure NME lesions (539 benign and 426 malignant) confirmed by histopathology or follow-up in 903 women. The 754 NME lesions acquired by one MR scanner were randomly split into the training set, validation set, and test set A (482/121/151 lesions). The 211 NME lesions acquired by another MR scanner were used as test set B. The AI system was developed using ResNet-50 with the axial and sagittal MIP images. One senior and one junior radiologist reviewed the MIP images of each case independently and rated its Breast Imaging Reporting and Data System category. The performance of the AI system and the radiologists was evaluated using the area under the receiver operating characteristic curve (AUC). RESULTS: The AI system yielded AUCs of 0.859 and 0.816 in the test sets A and B, respectively. The AI system achieved comparable performance as the senior radiologist (p = 0.558, p = 0.041) and outperformed the junior radiologist (p < 0.001, p = 0.009) in both test sets A and B. After AI assistance, the AUC of the junior radiologist increased from 0.740 to 0.862 in test set A (p < 0.001) and from 0.732 to 0.843 in test set B (p < 0.001). CONCLUSION: Our MIP-based AI system yielded good applicability in classifying NME lesions in breast MRI and can assist the junior radiologist achieve better performance. KEY POINTS: • Our MIP-based AI system yielded good applicability in the dataset both from the same and a different MR scanner in predicting malignant NME lesions. • The AI system achieved comparable diagnostic performance with the senior radiologist and outperformed the junior radiologist. • This AI system can assist the junior radiologist achieve better performance in the classification of NME lesions in MRI.

2.
Brain Imaging Behav ; 16(3): 1234-1245, 2022 Jun.
Artigo em Inglês | MEDLINE | ID: mdl-34973120

RESUMO

Tremor in Parkinson's disease (PD) has distinct responsiveness to dopamine, which is supposed not be exclusively related to dopamine deficiency but has a close relationship with cholinergic system. This phenomenon indicates that cholinergic system may be an important regulatory for distinct dopamine responsiveness of parkinsonian tremor. Through investigating the alterations of cholinergic and dopaminergic network during levodopa administration, we aimed at exploring the mechanisms of differed dopamine responsiveness of parkinsonian tremor. Fifty-two PD patients with tremor were enrolled. MRI scanning, UPDRS III and its sub-symptom scores were collected in OFF and ON status (dopaminergic challenge test). Then, patients were divided into two groups (dopamine-resistant tremor and dopamine-responsive tremor) according to the tremor change rate median score. Dopaminergic and cholinergic network were obtained. LASSO regression was conducted to identify functional connectivity with distinct reactivity during levodopa administration between groups. Afterwards, detailed group comparisons, interaction and correlation analyses were performed. The reactivity of cholinergic connectivity showed the highest possibility to distinguish two groups, especially connectivity of right basal forebrain 123 to right parietal operculum cortex (R.BF123-R.PO). After levodopa administration, connectivity of R.BF123-R.PO was decreased for dopamine-responsive tremor while which remained unchanged for dopamine-resistant tremor. The reactivity of R.BF123-R.PO was negatively correlated with tremor change rate. Reduced cholinergic connectivity to parietal operculum may be an underlying mechanism for the responsive tremor in PD and the distinct cholinergic reactivity of parietal operculum to levodopa may be a core pathophysiology for the differed DA responsiveness of tremor in PD.

3.
Acad Radiol ; 29 Suppl 3: S28-S35, 2022 03.
Artigo em Inglês | MEDLINE | ID: mdl-33160862

RESUMO

RATIONALE AND OBJECTIVE: To investigate age-related brain morphological changes of boys with high functioning autism (HFA). MATERIALS AND METHODS: Forty-six medication-naive boys with HFA and 48 age-matched typically developing boys (4-12 years old) were included in this study. Structural brain images were processed with FreeSurfer to calculate the brain morphometric features including regional volume, surface area, average cortical thickness, and Gaussian curvature. General linear model was used to identify significant effects of diagnosis and age-by-diagnosis interaction. Correlations between age and the brain morphometric variables of significant clusters were explored. RESULTS: Primarily, most of the regions with statistically significant intergroup differences were located in the temporal lobe gyri. Importantly, the volume of bilateral superior temporal gyrus (STG) and the average cortical thickness of the right STG demonstrated significantly age-related intergroup differences. Further age-stratified analysis also revealed morphological alterations of STG among subgroups of preschool and school-aged children with or without HFA. CONCLUSION: The findings demonstrated abnormal age-related volume and cortical thickness atrophy of the STG in HFA children, which reflect brain development trajectories of ASD may initiate to diverge from early overgrowth in childhood period. The anatomical localization of specific brain regions would help us better understand the neurobiology alterations of HFA patients and indicate the effect of age should be carefully delineated and examined in future studies about HFA.


Assuntos
Transtorno Autístico , Transtorno Autístico/diagnóstico por imagem , Encéfalo/diagnóstico por imagem , Mapeamento Encefálico/métodos , Criança , Pré-Escolar , Humanos , Imageamento por Ressonância Magnética/métodos , Masculino , Lobo Temporal
4.
Hum Brain Mapp ; 43(6): 1984-1996, 2022 04 15.
Artigo em Inglês | MEDLINE | ID: mdl-34970835

RESUMO

Identifying a whole-brain connectome-based predictive model in drug-naïve patients with Parkinson's disease and verifying its predictions on drug-managed patients would be useful in determining the intrinsic functional underpinnings of motor impairment and establishing general brain-behavior associations. In this study, we constructed a predictive model from the resting-state functional data of 47 drug-naïve patients by using a connectome-based approach. This model was subsequently validated in 115 drug-managed patients. The severity of motor impairment was assessed by calculating Unified Parkinson's Disease Rating Scale Part III scores. The predictive performance of model was evaluated using the correlation coefficient (rtrue ) between predicted and observed scores. As a result, a connectome-based model for predicting individual motor impairment in drug-naïve patients was identified with significant performance (rtrue  = .845, p < .001, ppermu  = .002). Two patterns of connection were identified according to correlations between connection strength and the severity of motor impairment. The negative motor-impairment-related network contained more within-network connections in the motor, visual-related, and default mode networks, whereas the positive motor-impairment-related network was constructed mostly with between-network connections coupling the motor-visual, motor-limbic, and motor-basal ganglia networks. Finally, this predictive model constructed around drug-naïve patients was confirmed with significant predictive efficacy on drug-managed patients (r = .209, p = .025), suggesting a generalizability in Parkinson's disease patients under long-term drug influence. In conclusion, this study identified a whole-brain connectome-based model that could predict the severity of motor impairment in Parkinson's patients and furthers our understanding of the functional underpinnings of the disease.


Assuntos
Conectoma , Transtornos Motores , Doença de Parkinson , Encéfalo/diagnóstico por imagem , Humanos , Imageamento por Ressonância Magnética , Doença de Parkinson/diagnóstico por imagem
5.
Front Oncol ; 11: 752158, 2021.
Artigo em Inglês | MEDLINE | ID: mdl-34745982

RESUMO

BACKGROUND: Brain invasion in meningioma has independent associations with increased risks of tumor progression, lesion recurrence, and poor prognosis. Therefore, this study aimed to construct a model for predicting brain invasion in WHO grade II meningioma by using preoperative MRI. METHODS: One hundred seventy-three patients with brain invasion and 111 patients without brain invasion were included. Three mainstream features, namely, traditional semantic features and radiomics features from tumor and tumor-to-brain interface regions, were acquired. Predictive models correspondingly constructed on each feature set or joint feature set were constructed. RESULTS: Traditional semantic findings, e.g., peritumoral edema and other four features, had comparable performance in predicting brain invasion with each radiomics feature set. By taking advantage of semantic features and radiomics features from tumoral and tumor-to-brain interface regions, an integrated nomogram that quantifies the risk factor of each selected feature was constructed and had the best performance in predicting brain invasion (area under the curve values were 0.905 in the training set and 0.895 in the test set). CONCLUSIONS: This study provided a clinically available and promising approach to predict brain invasion in WHO grade II meningiomas by using preoperative MRI.

6.
Eur Radiol ; 31(8): 5902-5912, 2021 Aug.
Artigo em Inglês | MEDLINE | ID: mdl-33496829

RESUMO

OBJECTIVES: To investigate the value of full-field digital mammography-based deep learning (DL) in predicting malignancy of Breast Imaging Reporting and Data System (BI-RADS) 4 microcalcifications. METHODS: A total of 384 patients with 414 pathologically confirmed microcalcifications (221 malignant and 193 benign) were randomly allocated into the training, validation, and testing datasets (272/71/71 lesions) in this retrospective study. A combined DL model was developed incorporating mammography and clinical variables. Model performance was evaluated by using areas under the receiver operating characteristic curve (AUC) and compared with the clinical model, stand-alone DL image model, and BI-RADS approach. The predictive performance for malignancy was also compared between the combined model and human readers (2 juniors and 2 seniors). RESULTS: The combined DL model demonstrated favorable AUC, sensitivity, and specificity of 0.910, 85.3%, and 91.9% in predicting BI-RADS 4 malignant microcalcifications in the testing dataset, which outperformed the clinical model, DL image model, and BI-RADS with AUCs of 0.799, 0.841, and 0.804, respectively. The combined model achieved non-inferior performance as senior radiologists (p = 0.860, p = 0.800) and outperformed junior radiologists (p = 0.155, p = 0.029). The diagnostic performance of two junior radiologists was improved after artificial intelligence assistance with AUCs increased to 0.854 and 0.901 from 0.816 (p = 0.556) and 0.773 (p = 0.046), while the interobserver agreement was improved with a kappa value increased to 0.843 from 0.331. CONCLUSIONS: The combined deep learning model can improve the malignancy prediction of BI-RADS 4 microcalcifications in screening mammography and assist junior radiologists to achieve better performance, which can facilitate clinical decision-making. KEY POINTS: • The combined deep learning model demonstrated high diagnostic power, sensitivity, and specificity for predicting malignant BI-RADS 4 mammographic microcalcifications. • The combined model achieved similar performance with senior breast radiologists, while it outperformed junior breast radiologists. • Deep learning could improve the diagnostic performance of junior radiologists and facilitate clinical decision-making.


Assuntos
Neoplasias da Mama , Calcinose , Aprendizado Profundo , Inteligência Artificial , Neoplasias da Mama/diagnóstico por imagem , Calcinose/diagnóstico por imagem , Detecção Precoce de Câncer , Feminino , Humanos , Mamografia , Estudos Retrospectivos
7.
Eur Radiol ; 31(5): 3080-3089, 2021 May.
Artigo em Inglês | MEDLINE | ID: mdl-33118047

RESUMO

OBJECTIVES: To construct a CT-based radiomics signature and assess its performance in predicting MYCN amplification (MNA) in pediatric patients with neuroblastoma. METHODS: Seventy-eight pediatric patients with neuroblastoma were recruited (55 in training cohort and 23 in test cohort). Radiomics features were extracted automatically from the region of interest (ROI) manually delineated on the three-phase computed tomography (CT) images. Selected radiomics features were retained to construct radiomics signature and a radiomics score (rad-score) was calculated by using the radiomics signature-based formula. A clinical model was established with clinical factors, including clinicopathological data, and CT image features. A combined nomogram was developed with the incorporation of a radiomics signature and clinical factors. The predictive performance was assessed by receiver operating characteristics curve (ROC) analysis and decision curve analysis (DCA). RESULTS: The radiomics signature was constructed using 7 selected radiomics features. The clinical radiomics nomogram, which was based on the radiomics signature and two clinical factors, showed superior predictive performance compared with the clinical model alone (area under the curve (AUC) in the training cohort: 0.95 vs. 0.82, the test cohort: 0.91 vs. 0.70). The clinical utility of clinical radiomics nomogram was confirmed by DCA. CONCLUSIONS: This proposed CT-based radiomics signature was able to predict MNA. Combining the radiomics signature with clinical factors outperformed using clinical model alone for MNA prediction. KEY POINTS: • A CT-based radiomics signature has the ability to predict MYCN amplification (MNA) in neuroblastoma. • Both pre- and post-contrast CT images are valuable in predicting MNA. • Associating the radiomics signature with clinical factors improved the predictive performance of MNA, compared with clinical model alone.


Assuntos
Neuroblastoma , Tomografia Computadorizada por Raios X , Criança , Humanos , Proteína Proto-Oncogênica N-Myc/genética , Neuroblastoma/diagnóstico por imagem , Neuroblastoma/genética , Nomogramas , Curva ROC
8.
Opt Express ; 28(20): 28831-28842, 2020 Sep 28.
Artigo em Inglês | MEDLINE | ID: mdl-33114793

RESUMO

A vector optical-chirp-chain (OCC) Brillouin optical time-domain analyzer (BOTDA) based on complex principal component analysis (CPCA) is proposed and experimentally demonstrated by employing a four-tone OCC probe with two orthogonal polarization states. The polarization-fading-free complex Brillouin spectrum (CBS) of the vector OCC-BOTDA is obtained by combining the amplitude and phase response spectra of the probe wave at both Brillouin gain and loss region. We utilize the CPCA method to determine the Brillouin frequency shift (BFS) directly using the measured CBS, and the sensing accuracy is improved by a factor of up to 1.4. The distributed temperature sensing is demonstrated over a 20 km standard single-mode fiber with a 6 m spatial resolution and less than 1 MHz frequency uncertainty under 10 times of trace averaging.

9.
Sensors (Basel) ; 20(5)2020 Feb 27.
Artigo em Inglês | MEDLINE | ID: mdl-32121020

RESUMO

The double tendon-sheath drive system is widely used in the design of surgical robots and search and rescue robots because of its simplicity, dexterity, and long-distance transmission. We are attempting to apply it to manipulators, wherenon-linear characteristics such as gaps, hysteresis, etc., due to friction between the contact surfaces of the tendon sheath and the flexibility of the rope, are the main difficulties in controlling such manipulators. Most of the existing compensation control methods applicable to double tendon-sheath actuators are offline compensation methods that do not require output feedback, but when the system's motion and configuration changes, it cannot adapt to the drastic changes in the transmission characteristics. Depending on the transmission system, the robotic arm, changes at any time during the working process, and the force sensors and torque sensors that cannot be applied to the joints of the robot, so a real-time position compensation control method based on flexible cable deformation is proposed. A double tendon-sheath transmission model is established, a double tendon-sheath torque transmission model under any load condition is derived, and a semi-physical simulation experimental platform composed of a motor, a double tendon-sheath transmission system and a single articulated arm is established to verify the transfer model. Through the signal feedback of the end encoder, a real-time closed-loop feedback system was established, thus that the system can still achieve the output to follow the desired torque trajectory under the external interference.


Assuntos
Procedimentos Cirúrgicos Robóticos/métodos , Algoritmos , Sistemas Computacionais , Desenho de Equipamento , Torque
10.
Hell J Nucl Med ; 22(1): 49-57, 2019.
Artigo em Inglês | MEDLINE | ID: mdl-30843010

RESUMO

OBJECTIVE: An indigenous polymethyl metacrylate (PMMA) phantom with a V-shaped slit and a correlated technique for semi-quantifying the minimum detectable difference (MDD) of single photon emission tomography (SPET) via gamma camera scanning are proposed and validated using four radionuclides. MATERIALS AND METHODS: Radio-actinide solutions of gallium-67 (67Ga), technetium-99m (99mTc), iodine-131 (131I) and thallium-201 (201Tl) were diluted to 11c.c. and thoroughly injected into the continuous zig zag slit of the PMMA phantom. Either depth or edge of the slit between two lines of the V-shape was customized from deep or wide to change into shallow or narrow gradually. Thus, the quantified MDD could be easily evaluated, according to the revised Student's t-test evaluation. The revised Student's t-test was calculated by both full width at half maximum (FWHM) and edge width between two adjacent peaks that were acquired from the original data matrix of SPET. The derived MDD was indicated as for radionuclide, depth, width in mm: For 67Ga, 2.9, 2.13, for 99mTc, 2.5, 0.66, for 131I, 4.7, 2.38 and for 201Tl, 3.3, 2.00, respectively. RESULTS: Technetium-99m had the highest and 131I had the lowest MDD among the four radionuclides. Furthermore, two adjacent peaks of 67Ga could be easily identified with fewer counts than for 201Tl (depth, 2.9 vs. 3.3mm), but its MDD was poorer (width: 2.13 vs.2.00mm). The revised Student's t-test analysis proved to be an acceptable technique for the MDD identification. CONCLUSION: The proposed new combination of PMMA phantom with a V-slit and the revised Student's t-test proved to be instrumental in the MDD of SPET optimization analysis.


Assuntos
Limite de Detecção , Imagens de Fantasmas/normas , Compostos Radiofarmacêuticos , Tomografia Computadorizada de Emissão de Fóton Único/instrumentação , Câmaras gama/normas , Humanos , Radioisótopos do Iodo , Polimetil Metacrilato , Tecnécio , Tálio , Tomografia Computadorizada de Emissão de Fóton Único/métodos , Tomografia Computadorizada de Emissão de Fóton Único/normas
SELEÇÃO DE REFERÊNCIAS
DETALHE DA PESQUISA
...