Your browser doesn't support javascript.
loading
Mostrar: 20 | 50 | 100
Resultados 1 - 20 de 56
Filtrar
Mais filtros

Base de dados
País/Região como assunto
Tipo de documento
País de afiliação
Intervalo de ano de publicação
1.
Semin Cancer Biol ; 96: 11-25, 2023 11.
Artigo em Inglês | MEDLINE | ID: mdl-37704183

RESUMO

Breast cancer is a significant global health burden, with increasing morbidity and mortality worldwide. Early screening and accurate diagnosis are crucial for improving prognosis. Radiographic imaging modalities such as digital mammography (DM), digital breast tomosynthesis (DBT), magnetic resonance imaging (MRI), ultrasound (US), and nuclear medicine techniques, are commonly used for breast cancer assessment. And histopathology (HP) serves as the gold standard for confirming malignancy. Artificial intelligence (AI) technologies show great potential for quantitative representation of medical images to effectively assist in segmentation, diagnosis, and prognosis of breast cancer. In this review, we overview the recent advancements of AI technologies for breast cancer, including 1) improving image quality by data augmentation, 2) fast detection and segmentation of breast lesions and diagnosis of malignancy, 3) biological characterization of the cancer such as staging and subtyping by AI-based classification technologies, 4) prediction of clinical outcomes such as metastasis, treatment response, and survival by integrating multi-omics data. Then, we then summarize large-scale databases available to help train robust, generalizable, and reproducible deep learning models. Furthermore, we conclude the challenges faced by AI in real-world applications, including data curating, model interpretability, and practice regulations. Besides, we expect that clinical implementation of AI will provide important guidance for the patient-tailored management.


Assuntos
Neoplasias da Mama , Humanos , Feminino , Neoplasias da Mama/diagnóstico por imagem , Inteligência Artificial , Prognóstico , Mamografia , Multiômica , Mama
2.
Mol Breed ; 44(4): 28, 2024 Apr.
Artigo em Inglês | MEDLINE | ID: mdl-38545461

RESUMO

Powdery mildew, caused by Blumeria graminis f. sp. tritici (Bgt), is a severe disease that affects the yield and quality of wheat. Popularization of resistant cultivars in production is the preferred strategy to control this disease. In the present study, the Chinese wheat breeding line Jimai 809 showed excellent agronomic performance and high resistance to powdery mildew at the whole growth stage. To dissect the genetic basis for this resistance, Jimai 809 was crossed with the susceptible wheat cultivar Junda 159 to produce segregation populations. Genetic analysis showed that a single dominant gene, temporarily designated PmJM809, conferred the resistance to different Bgt isolates. PmJM809 was then mapped on the chromosome arm 2BL and flanked by the markers CISSR02g-1 and CIT02g-13 with genetic distances 0.4 and 0.8 cM, respectively, corresponding to a physical interval of 704.12-708.24 Mb. PmJM809 differed from the reported Pm genes on chromosome arm 2BL in origin, resistance spectrum, physical position and/or genetic diversity of the mapping interval, also suggesting PmJM809 was located on a complex interval with multiple resistance genes. To analyze and screen the candidate gene(s) of PmJM809, six genes related to disease resistance in the candidate interval were evaluated their expression patterns using an additional set of wheat samples and time-course analysis post-inoculation of the Bgt isolate E09. As a result, four genes were speculated as the key candidate or regulatory genes. Considering its comprehensive agronomic traits and resistance findings, PmJM809 was expected to be a valuable gene resource in wheat disease resistance breeding. To efficiently transfer PmJM809 into different genetic backgrounds, 13 of 19 closely linked markers were confirmed to be suitable for marker-assisted selection. Using these markers, a series of wheat breeding lines with harmonious disease resistance and agronomic performance were selected from the crosses of Jimai 809 and several susceptible cultivars. Supplementary Information: The online version contains supplementary material available at 10.1007/s11032-024-01467-8.

3.
Rev Cardiovasc Med ; 24(1): 7, 2023 Jan.
Artigo em Inglês | MEDLINE | ID: mdl-39076877

RESUMO

Background: Hypoperfusion, a common manifestation of many critical illnesses, could lead to abnormalities in body surface thermal distribution. However, the interpretation of thermal images is difficult. Our aim was to assess the mortality risk of critically ill patients at risk of hypoperfusion in a prospective cohort by infrared thermography combined with deep learning methods. Methods: This post-hoc study was based on a cohort at high-risk of hypoperfusion. Patients' legs were selected as the region of interest. Thermal images and conventional hypoperfusion parameters were collected. Six deep learning models were attempted to derive the risk of mortality (range: 0 to 100%) for each patient. The area under the receiver operating characteristic curve (AUROC) was used to evaluate predictive accuracy. Results: Fifty-five hospital deaths occurred in a cohort consisting of 373 patients. The conventional hypoperfusion (capillary refill time and diastolic blood pressure) and thermal (low temperature area rate and standard deviation) parameters demonstrated similar predictive accuracies for hospital mortality (AUROC 0.73 and 0.77). The deep learning methods, especially the ResNet (18), could further improve the accuracy. The AUROC of ResNet (18) was 0.94 with a sensitivity of 84% and a specificity of 91% when using a cutoff of 36%. ResNet (18) presented a significantly increasing trend in the risk of mortality in patients with normotension (13 [7 to 26]), hypotension (18 [8 to 32]) and shock (28 [14 to 62]). Conclusions: Interpreting infrared thermography with deep learning enables accurate and non-invasive assessment of the severity of patients at risk of hypoperfusion.

4.
Mikrochim Acta ; 190(12): 466, 2023 11 13.
Artigo em Inglês | MEDLINE | ID: mdl-37953315

RESUMO

The successful development of a dual-mode sensing chip for deoxynivalenol (DON) detection using photoelectrochemical (PEC) and electrochromic visualization techniques is reported. By laser etching technology, different functional areas, including the photoanode, the cathode, and the electrochromic area, are fabricated on indium tin oxide (ITO) glass. Then, these three areas are further respectively modified with PEC active materials, platinum nanoparticles, and Prussian blue. Under light illumination, photocurrents generate between the photoanode and the cathode due to the separation of photo-induced electrons and holes in the TiO2/3DNGH material. Meanwhile, the photo-induced electrons are transferred to Prussian blue on the visualization area, which will be reduced to colorless Prussian white. The binding of DON molecules and aptamers can promote electron transfer and reduce the recombination of electrons and holes, allowing for simultaneous quantitative detection of DON using either the photocurrent or color change. The sensor chip has a broad detection range of DON concentrations of 1 fg⋅mL-1 to 100 pg⋅mL-1 in the PEC mode with the limit of detection of 0.37 fg⋅mL-1, and 1 to 250 ng⋅mL-1 in the visualization mode with the limit of detection of 0.51 ng⋅mL-1. This portable dual-mode sensor chip can be used in both laboratory and field settings without the need for specialized instruments, making it a powerful tool for ensuring food safety. At the same time, the analysis of the standard addition method of the actual sample by using the sensor chip shows that, in the PEC mode, the recoveries of the dual-mode aptasensor chip were 91.3 to 99.0% with RSD values of 1.73~2.55%, and in visualization mode, the recoveries of the dual-mode aptasensor chip were 99.2 to 102.0% with RSD values of 1.00~6.21%, which indicate good accuracy and reproducibility.


Assuntos
Técnicas Biossensoriais , Nanopartículas Metálicas , Nanopartículas Metálicas/química , Reprodutibilidade dos Testes , Platina
5.
Sensors (Basel) ; 23(23)2023 Nov 26.
Artigo em Inglês | MEDLINE | ID: mdl-38067788

RESUMO

Active mapping is an important technique for mobile robots to autonomously explore and recognize indoor environments. View planning, as the core of active mapping, determines the quality of the map and the efficiency of exploration. However, most current view-planning methods focus on low-level geometric information like point clouds and neglect the indoor objects that are important for human-robot interaction. We propose a novel View-Planning method for indoor active Sparse Object Mapping (VP-SOM). VP-SOM takes into account for the first time the properties of object clusters in the coexisting human-robot environment. We categorized the views into global views and local views based on the object cluster, to balance the efficiency of exploration and the mapping accuracy. We developed a new view-evaluation function based on objects' information abundance and observation continuity, to select the Next-Best View (NBV). Especially for calculating the uncertainty of the sparse object model, we built the object surface occupancy probability map. Our experimental results demonstrated that our view-planning method can explore the indoor environments and build object maps more accurately, efficiently, and robustly.

7.
Tumour Biol ; 36(10): 7961-6, 2015 Sep.
Artigo em Inglês | MEDLINE | ID: mdl-25957892

RESUMO

Although dendritic cells (DCs) used in DC-based immunotherapy are loaded with tumor-associated antigens, the antitumor immune response is effectively stimulated in cytotoxic specific T lymphocytes (CTLs), thereby achieving targeted killing of tumor cells without harming surrounding normal cells. This makes it a highly promising new form of therapy. In this study, we successfully created chitosan-coated EphrinA1-PE38/GM-CSF nanoparticles and transplanted them into tumor-bearing rats, resulting in the effective killing of glioma tissue and a prolonged life span. Further immunohistochemistry and flow cytometry studies revealed that the treatment was effective in increasing the number of dendritic cell activations under an immunomodulatory response. These results suggest that chitosan-coated EphrinA1-PE38/GM-CSF nanoparticles may be effective in inducing in situ activation of DCs in glioma-bearing rats, thereby generating DC vaccines in vivo.


Assuntos
Quitosana/imunologia , Células Dendríticas/imunologia , Efrina-A1/imunologia , Glioma/terapia , Fator Estimulador de Colônias de Granulócitos e Macrófagos/imunologia , Imunoterapia , Nanopartículas/administração & dosagem , Animais , Neoplasias Encefálicas/imunologia , Neoplasias Encefálicas/patologia , Neoplasias Encefálicas/terapia , Citometria de Fluxo , Imunofluorescência , Glioma/imunologia , Glioma/patologia , Técnicas Imunoenzimáticas , Ratos , Linfócitos T Citotóxicos/imunologia , Células Tumorais Cultivadas
8.
Tumour Biol ; 36(7): 5497-503, 2015 Jul.
Artigo em Inglês | MEDLINE | ID: mdl-25677907

RESUMO

Dendritic cells loaded with tumor-associated antigens can effectively stimulate the antitumor immune response of cytotoxic T lymphocytes in the body, which facilitates the development of novel and effective treatments for cancer. In this study, the adenovirus-mediated ephrinA1-PE38/GM-CSF was successfully constructed using the overlap extension method, and verified with sequencing analysis. HEK293 cells were infected with the adenovirus and the cellular expression of ephrinA1-PE38/GM-CSF was measured with an enzyme-linked immunosorbent assay. The recombinant adenovirus was then delivered into the tumor-bearing rats and the results showed that such treatment significantly reduced the volumes of gliomas and improved the survival of the transplanted rats. The results from immunohistochemistry and flow cytometry suggested that this immunomodulatory agent cause activation of dendritic cells. The findings that ephrinA1-PE38/GM-CSF had a high efficacy in the activation of the dendritic cells would facilitate the development of in vivo dendritic-cell vaccines for the treatment of gliomas in rats. Our new method of DC vaccine production induces not only a specific local antitumor immune response but also a systemic immunotherapeutic effect. In addition, this method completely circumvents the risk of contamination related to the in vitro culture of DCs, thus greatly improving the safety and feasibility of clinical application of the DC vaccines in glioma.


Assuntos
ADP Ribose Transferases/imunologia , Toxinas Bacterianas/imunologia , Vacinas Anticâncer/imunologia , Efrina-A1/imunologia , Exotoxinas/imunologia , Glioma/imunologia , Fator Estimulador de Colônias de Granulócitos e Macrófagos/imunologia , Proteínas Recombinantes/genética , Fatores de Virulência/imunologia , ADP Ribose Transferases/genética , Adenoviridae/genética , Animais , Antígenos de Neoplasias/imunologia , Toxinas Bacterianas/genética , Vacinas Anticâncer/genética , Células Dendríticas/imunologia , Efrina-A1/genética , Exotoxinas/genética , Glioma/genética , Glioma/prevenção & controle , Fator Estimulador de Colônias de Granulócitos e Macrófagos/genética , Células HEK293 , Humanos , Imunidade Celular/genética , Imunidade Celular/imunologia , Imunomodulação , Ratos , Proteínas Recombinantes/administração & dosagem , Linfócitos T Citotóxicos/imunologia , Fatores de Virulência/genética , Exotoxina A de Pseudomonas aeruginosa
9.
J Stroke Cerebrovasc Dis ; 23(10): 2591-2597, 2014.
Artigo em Inglês | MEDLINE | ID: mdl-25267587

RESUMO

This study aimed to investigate the combination effects of bone marrow stromal cells (BMSCs) and oxiracetam for ischemic stroke. Forty Sprague Dawley female rats (220 ± 20 g) were subjected to a 2-hour ischemic middle cerebral artery occlusion (MCAO)-24 hours reperfusion model. The rats were randomly divided into 4 groups. Rats from BMSCs group, oxiracetam group, and BMSCs + oxiracetam group accepted injection of BMSCs (3 × 10(6) cells), oxiracetam (800 mg/kg), and BMSCs + oxiracetam, respectively. Rats from control group did not receive any interventions after ischemia reperfusion. The neurologic function was examined by modified neurological severity scores (mNSS). B-cell lymphoma 2 (Bcl-2) expression and apoptosis were detected by immunohistochemistry and terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) staining. The mNSS was decreased in all treatment groups and that in BMSCs + oxiracetam group was lower than BMSCs group and oxiracetam group (P < .05). The expression of Bcl-2 was unregulated in all treatment groups (P < .05), and similarly, the expression of Bcl-2 in BMSCs + oxiracetam group was higher than BMSCs group and oxiracetam group (P < .05). Control group displayed more TUNEL-positive cells than the treatment groups, and BMSCs + oxiracetam group displayed less apoptotic cells than BMSCs group or oxiracetam group (P < .05). Transplantation of BMSCs can promote the recovery of neurologic function in MCAO rats, and the effect of BMSCs combined with oxiracetam was better than the either one. Upregulation of Bcl-2 resulting in a decrease of apoptosis may be one of the mechanisms of BMSCs treatment for cerebral ischemic stroke.


Assuntos
Células da Medula Óssea/citologia , Transplante de Medula Óssea/métodos , Isquemia Encefálica/terapia , Células-Tronco Mesenquimais/metabolismo , Pirrolidinas/farmacologia , Acidente Vascular Cerebral/terapia , Animais , Células da Medula Óssea/metabolismo , Isquemia Encefálica/metabolismo , Terapia Combinada/métodos , Modelos Animais de Doenças , Relação Dose-Resposta a Droga , Feminino , Células-Tronco Mesenquimais/citologia , Nootrópicos/farmacologia , Proteínas Proto-Oncogênicas c-bcl-2/metabolismo , Pirrolidinas/administração & dosagem , Ratos , Ratos Sprague-Dawley , Acidente Vascular Cerebral/metabolismo , Resultado do Tratamento , Regulação para Cima/efeitos dos fármacos
10.
Zhonghua Yi Xue Za Zhi ; 94(29): 2286-9, 2014 Aug 05.
Artigo em Zh | MEDLINE | ID: mdl-25391873

RESUMO

OBJECTIVE: To explore the individual surgical treatment of various performance types of craniocervical junction malformation. METHODS: From January 2011 to December 2013, 112 patients with craniocervical junction malformations were treated at our department, including Chiari malformation (n = 65) (syringomyelia, n = 58 and without syringomyelia, n = 7), basilar invagination disease (n = 35) (with cerebellar tonsillar herniation malformation or occipitocervical fusion) and complex craniocervical malformation (n = 22) (atlantoaxial dislocation with occipitocervical fusion or with chiari malformation or cervical insufficiency sub-section). All of them had the symptoms of upper cervical nerve damage. For those with Chiari malformation, we evaluated atlanto-occipital joint stability preoperatively. If atlanto-occipital joint was stable, we performed small occipital bone window decompression, partial removal of cerebellar tonsils, loosening of posterior fossa, upper cervical adhesions, artificial dura appropriate sutured dural repair expanding neck pillow. For patients with basilar invagination, if nerve compression performance was in the rear, posterior decompression was performed. For those with complex craniocervical malformation with atlantoaxial dislocation, neck traction under anesthesia or traction after anterior release, then pillow neck fixation and fusion were performed. RESULTS: During follow-ups, the symptoms improved significantly (n = 98, 87.51%). There were no symptomatic change (n = 10, 8.93%), postoperative neurological deterioration (n = 3, 2.67%) and death (n = 1, 0.89%). CONCLUSION: According to specific clinical manifestations of craniocervical junction malformation patients, the best treatment is to perform individualized surgeries after thorough preoperative evaluations.


Assuntos
Anormalidades Craniofaciais/cirurgia , Adulto , Descompressão Cirúrgica , Dura-Máter , Feminino , Humanos , Masculino , Período Pós-Operatório , Fusão Vertebral , Tração
11.
World J Clin Cases ; 12(11): 1990-1995, 2024 Apr 16.
Artigo em Inglês | MEDLINE | ID: mdl-38660553

RESUMO

BACKGROUND: When an anorectal foreign body is found, its composition and shape should be evaluated, and a timely and effective treatment plan should be developed based on the patient's symptoms to avoid serious complications such as intestinal perforation caused by displacement of the foreign body. CASE SUMMARY: A 54-year-old male was admitted to our outpatient clinic on June 3, 2023, due to a rectal foreign body that had been embedded for more than 24 h. The patient reported using a glass electrode tube to assist in the recovery of prolapsed hemorrhoids, however, the electrode tube was inadvertently inserted into the anus and could not be removed by the patient. During hospitalization, the patient underwent surgery, and the foreign body was dragged into the rectum with the aid of colonoscopy. The anus was dilated with a comb-type pulling hook and an anal fistula pulling hook to widen the anus and remove the foreign body, and the local anal symptoms were then relieved with topical drugs. The patient was allowed to eat and drink, and an entire abdominal Computed tomography (CT) and colonoscopy were reviewed 3 d after surgery. CT revealed no foreign body residue and colonoscopy showed no metal or other residues in the colon and rectum, and no apparent intestinal tract damage. CONCLUSION: The timeliness and rationality of the surgical and therapeutic options for this patient were based on a literature review of the clinical signs and conceivable conditions in such cases. The type, material and the potential risks of rectal foreign bodies should be considered.

12.
Med Image Anal ; 92: 103045, 2024 Feb.
Artigo em Inglês | MEDLINE | ID: mdl-38071865

RESUMO

Automatic and accurate dose distribution prediction plays an important role in radiotherapy plan. Although previous methods can provide promising performance, most methods did not consider beam-shaped radiation of treatment delivery in clinical practice. This leads to inaccurate prediction, especially on beam paths. To solve this problem, we propose a beam-wise dose composition learning (BDCL) method for dose prediction in the context of head and neck (H&N) radiotherapy plan. Specifically, a global dose network is first utilized to predict coarse dose values in the whole-image space. Then, we propose to generate individual beam masks to decompose the coarse dose distribution into multiple field doses, called beam voters, which are further refined by a subsequent beam dose network and reassembled to form the final dose distribution. In particular, we design an overlap consistency module to keep the similarity of high-level features in overlapping regions between different beam voters. To make the predicted dose distribution more consistent with the real radiotherapy plan, we also propose a dose-volume histogram (DVH) calibration process to facilitate feature learning in some clinically concerned regions. We further apply an edge enhancement procedure to enhance the learning of the extracted feature from the dose falloff regions. Experimental results on a public H&N cancer dataset from the AAPM OpenKBP challenge show that our method achieves superior performance over other state-of-the-art approaches by significant margins. Source code is released at https://github.com/TL9792/BDCLDosePrediction.


Assuntos
Neoplasias de Cabeça e Pescoço , Radioterapia de Intensidade Modulada , Humanos , Dosagem Radioterapêutica , Planejamento da Radioterapia Assistida por Computador/métodos , Radioterapia de Intensidade Modulada/métodos , Neoplasias de Cabeça e Pescoço/radioterapia
13.
Med Phys ; 51(4): 2678-2694, 2024 Apr.
Artigo em Inglês | MEDLINE | ID: mdl-37862556

RESUMO

BACKGROUND: Ovarian cancer is a highly lethal gynecological disease. Accurate and automated segmentation of ovarian tumors in contrast-enhanced computed tomography (CECT) images is crucial in the radiotherapy treatment of ovarian cancer, enabling radiologists to evaluate cancer progression and develop timely therapeutic plans. However, automatic ovarian tumor segmentation is challenging due to factors such as inhomogeneous background, ambiguous tumor boundaries, and imbalanced foreground-background, all of which contribute to high predictive uncertainty for a segmentation model. PURPOSE: To tackle these challenges, we propose an uncertainty-aware refinement framework that aims to estimate and refine regions with high predictive uncertainty for accurate ovarian tumor segmentation in CECT images. METHODS: To this end, we first employ an approximate Bayesian network to detect coarse regions of interest (ROIs) of both ovarian tumors and uncertain regions. These ROIs allow a subsequent segmentation network to narrow down the search area for tumors and prioritize uncertain regions, resulting in precise segmentation of ovarian tumors. Meanwhile, the framework integrates two guidance modules that learn two implicit functions capable of mapping query features sampled according to their uncertainty to organ or boundary manifolds, guiding the segmentation network to facilitate information encoding of uncertain regions. RESULTS: Firstly, 367 CECT images are collected from the same hospital for experiments. Dice score, Jaccard, Recall, Positive predictive value (PPV), 95% Hausdorff distance (HD95) and Average symmetric surface distance (ASSD) for the testing group of 77 cases are 86.31%, 73.93%, 83.95%, 86.03%, 15.17  mm and 2.57  mm, all of which are significantly better than that of the other state-of-the-art models. And results of visual comparison shows that the compared methods have more mis-segmentation than our method. Furthermore, our method achieves a Dice score that is at least 20% higher than the Dice scores of other compared methods when tumor volumes are less than 20 cm 3 $^3$ , indicating better recognition ability to small regions by our method. And then, 38 CECT images are collected from another hospital to form an external testing group. Our approach consistently outperform the compared methods significantly, with the external testing group exhibiting substantial improvements across key evaluation metrics: Dice score (83.74%), Jaccard (69.55%), Recall (82.12%), PPV (81.61%), HD95 (12.31 mm), and ASSD (2.32 mm), robustly establishing its superior performance. CONCLUSIONS: Experimental results demonstrate that the framework significantly outperforms the compared state-of-the-art methods, with decreased under- or over-segmentation and better small tumor identification. It has the potential for clinical application.


Assuntos
Neoplasias Ovarianas , Feminino , Humanos , Teorema de Bayes , Incerteza , Neoplasias Ovarianas/diagnóstico por imagem , Aprendizagem , Benchmarking , Processamento de Imagem Assistida por Computador
14.
Food Chem ; 458: 140231, 2024 Jun 25.
Artigo em Inglês | MEDLINE | ID: mdl-38959803

RESUMO

Aflatoxin B1 (AFB1), a pernicious constituent of the aflatoxin family, predominantly contaminates cereals, oils, and their derivatives. Acknowledged as a Class I carcinogen by the World Health Organization (WHO), the expeditious and quantitative discernment of AFB1 remains imperative. This investigation delineates that aluminum ions can precipitate the coalescence of iodine-modified silver nanoparticles, thereby engendering hot spots conducive for label-free AFB1 identification via Surface-Enhanced Raman Spectroscopy (SERS). This methodology manifests a remarkable limit of detection (LOD) at 0.47 fg/mL, surpassing the sensitivity thresholds of conventional survey techniques. Moreover, this method has good anti-interference ability, with a relative error of less than 10% and a relative standard deviation of less than 6% in quantitative results. Collectively, these findings illuminate the substantial application potential and viability of this approach in the quantitative analysis of AFB1, underpinning a significant advancement in food safety diagnostics.

15.
IEEE Trans Med Imaging ; PP2024 Jul 08.
Artigo em Inglês | MEDLINE | ID: mdl-38976464

RESUMO

Medical image analysis poses significant challenges due to limited availability of clinical data, which is crucial for training accurate models. This limitation is further compounded by the specialized and labor-intensive nature of the data annotation process. For example, despite the popularity of computed tomography angiography (CTA) in diagnosing atherosclerosis with an abundance of annotated datasets, magnetic resonance (MR) images stand out with better visualization for soft plaque and vessel wall characterization. However, the higher cost and limited accessibility of MR, as well as time-consuming nature of manual labeling, contribute to fewer annotated datasets. To address these issues, we formulate a multi-modal transfer learning network, named MT-Net, designed to learn from unpaired CTA and sparsely-annotated MR data. Additionally, we harness the Segment Anything Model (SAM) to synthesize additional MR annotations, enriching the training process. Specifically, our method first segments vessel lumen regions followed by precise characterization of carotid artery vessel walls, thereby ensuring both segmentation accuracy and clinical relevance. Validation of our method involved rigorous experimentation on publicly available datasets from COSMOS and CARE-II challenge, demonstrating its superior performance compared to existing state-of-the-art techniques.

16.
Eur J Radiol ; 174: 111402, 2024 May.
Artigo em Inglês | MEDLINE | ID: mdl-38461737

RESUMO

PURPOSE: To assess the feasibility and clinical value of synthetic diffusion kurtosis imaging (DKI) generated from diffusion weighted imaging (DWI) through multi-task reconstruction network (MTR-Net) for tumor response prediction in patients with locally advanced rectal cancer (LARC). METHODS: In this retrospective study, 120 eligible patients with LARC were enrolled and randomly divided into training and testing datasets with a 7:3 ratio. The MTR-Net was developed for reconstructing Dapp and Kapp images from apparent diffusion coefficient (ADC) images. Tumor regions were manually segmented on both true and synthetic DKI images. The synthetic image quality and manual segmentation agreement were quantitatively assessed. The support vector machine (SVM) classifier was used to construct radiomics models based on the true and synthetic DKI images for pathological complete response (pCR) prediction. The prediction performance for the models was evaluated by the receiver operating characteristic (ROC) curve analysis. RESULTS: The mean squared error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM) for tumor regions were 0.212, 24.278, and 0.853, respectively, for the synthetic Dapp images and 0.516, 24.883, and 0.804, respectively, for the synthetic Kapp images. The Dice similarity coefficient (DSC), positive predictive value (PPV), sensitivity (SEN), and Hausdorff distance (HD) for the manually segmented tumor regions were 0.786, 0.844, 0.755, and 0.582, respectively. For predicting pCR, the true and synthetic DKI-based radiomics models achieved area under the curve (AUC) values of 0.825 and 0.807 in the testing datasets, respectively. CONCLUSIONS: Generating synthetic DKI images from DWI images using MTR-Net is feasible, and the efficiency of synthetic DKI images in predicting pCR is comparable to that of true DKI images.


Assuntos
Segunda Neoplasia Primária , Neoplasias Retais , Humanos , Estudos Retrospectivos , Terapia Neoadjuvante , Imagem de Difusão por Ressonância Magnética/métodos , Neoplasias Retais/diagnóstico por imagem , Neoplasias Retais/terapia , Neoplasias Retais/patologia , Quimiorradioterapia
17.
IEEE Trans Med Imaging ; 43(5): 1958-1971, 2024 May.
Artigo em Inglês | MEDLINE | ID: mdl-38206779

RESUMO

Breast cancer is becoming a significant global health challenge, with millions of fatalities annually. Magnetic Resonance Imaging (MRI) can provide various sequences for characterizing tumor morphology and internal patterns, and becomes an effective tool for detection and diagnosis of breast tumors. However, previous deep-learning based tumor segmentation methods from multi-parametric MRI still have limitations in exploring inter-modality information and focusing task-informative modality/modalities. To address these shortcomings, we propose a Modality-Specific Information Disentanglement (MoSID) framework to extract both inter- and intra-modality attention maps as prior knowledge for guiding tumor segmentation. Specifically, by disentangling modality-specific information, the MoSID framework provides complementary clues for the segmentation task, by generating modality-specific attention maps to guide modality selection and inter-modality evaluation. Our experiments on two 3D breast datasets and one 2D prostate dataset demonstrate that the MoSID framework outperforms other state-of-the-art multi-modality segmentation methods, even in the cases of missing modalities. Based on the segmented lesions, we further train a classifier to predict the patients' response to radiotherapy. The prediction accuracy is comparable to the case of using manually-segmented tumors for treatment outcome prediction, indicating the robustness and effectiveness of the proposed segmentation method. The code is available at https://github.com/Qianqian-Chen/MoSID.


Assuntos
Neoplasias da Mama , Interpretação de Imagem Assistida por Computador , Imageamento por Ressonância Magnética , Humanos , Neoplasias da Mama/diagnóstico por imagem , Feminino , Interpretação de Imagem Assistida por Computador/métodos , Imageamento por Ressonância Magnética/métodos , Masculino , Algoritmos , Aprendizado Profundo , Mama/diagnóstico por imagem , Bases de Dados Factuais , Neoplasias da Próstata/diagnóstico por imagem
18.
Ultrasound Med Biol ; 50(1): 18-27, 2024 01.
Artigo em Inglês | MEDLINE | ID: mdl-37806923

RESUMO

OBJECTIVE: Photoacoustic imaging has undergone rapid development in recent years. To simulate photoacoustic imaging on a computer, the most popular MATLAB toolbox currently used for the forward projection process is k-Wave. However, k-Wave suffers from significant computation time. Here we propose a straightforward simulation approach based on superposed Wave (s-Wave) to accelerate photoacoustic simulation. METHODS: In this study, we consider the initial pressure distribution as a collection of individual pixels. By obtaining standard sensor data from a single pixel beforehand, we can easily manipulate the phase and amplitude of the sensor data for specific pixels using loop and multiplication operators. The effectiveness of this approach is validated through an optimization-based reconstruction algorithm. RESULTS: The results reveal significantly reduced computation time compared with k-Wave. Particularly in a sparse 3-D configuration, s-Wave exhibits a speed improvement >2000 times compared with k-Wave. In terms of optimization-based image reconstruction, in vivo imaging results reveal that using the s-Wave method yields images highly similar to those obtained using k-Wave, while reducing the reconstruction time by approximately 50 times. CONCLUSION: Proposed here is an accelerated optimization-based algorithm for photoacoustic image reconstruction, using the fast s-Wave forward projection simulation. Our method achieves substantial time savings, particularly in sparse system configurations. Future work will focus on further optimizing the algorithm and expanding its applicability to a broader range of photoacoustic imaging scenarios.


Assuntos
Algoritmos , Técnicas Fotoacústicas , Imagens de Fantasmas , Simulação por Computador , Análise Espectral , Processamento de Imagem Assistida por Computador/métodos , Técnicas Fotoacústicas/métodos
19.
IEEE Trans Med Imaging ; PP2024 Jul 29.
Artigo em Inglês | MEDLINE | ID: mdl-39074000

RESUMO

Automated breast tumor segmentation on the basis of dynamic contrast-enhancement magnetic resonance imaging (DCE-MRI) has shown great promise in clinical practice, particularly for identifying the presence of breast disease. However, accurate segmentation of breast tumor is a challenging task, often necessitating the development of complex networks. To strike an optimal tradeoff between computational costs and segmentation performance, we propose a hybrid network via the combination of convolution neural network (CNN) and transformer layers. Specifically, the hybrid network consists of a encoder-decoder architecture by stacking convolution and deconvolution layers. Effective 3D transformer layers are then implemented after the encoder subnetworks, to capture global dependencies between the bottleneck features. To improve the efficiency of hybrid network, two parallel encoder sub-networks are designed for the decoder and the transformer layers, respectively. To further enhance the discriminative capability of hybrid network, a prototype learning guided prediction module is proposed, where the category-specified prototypical features are calculated through online clustering. All learned prototypical features are finally combined with the features from decoder for tumor mask prediction. The experimental results on private and public DCE-MRI datasets demonstrate that the proposed hybrid network achieves superior performance than the state-of-the-art (SOTA) methods, while maintaining balance between segmentation accuracy and computation cost. Moreover, we demonstrate that automatically generated tumor masks can be effectively applied to identify HER2-positive subtype from HER2-negative subtype with the similar accuracy to the analysis based on manual tumor segmentation. The source code is available at https://github.com/ZhouL-lab/ PLHN.

20.
Biosens Bioelectron ; 260: 116455, 2024 Sep 15.
Artigo em Inglês | MEDLINE | ID: mdl-38824702

RESUMO

In this work, a potential-controlled electrochromic visual biosensor was developed for detecting zearalenone (ZEN) using a distance readout strategy. The sensor chip includes a square detection area and a folded signal output area created with laser etching technology. The detection area is modified with graphene oxide and ZEN aptamer, while Prussian blue (PB) is electrodeposited onto the signal output channel. When an appropriate voltage is applied, PB in the signal output area is reduced to colorless Prussian white (PW). The target ZEN molecules have the capability to release aptamers from graphene oxide (GO) surface in the detection area, resulting in a subsequent change in the potential of the visual signal output channel. This change determines the length of the channel that changes from blue to colorless, with the color change distance being proportional to the ZEN concentration. Using this distance readout strategy, ZEN detection within the range of 1 ng/mL to 300 ng/mL was achieved, with a detection limit of 0.29 ng/mL.


Assuntos
Aptâmeros de Nucleotídeos , Técnicas Biossensoriais , Grafite , Limite de Detecção , Zearalenona , Zearalenona/análise , Técnicas Biossensoriais/métodos , Técnicas Biossensoriais/instrumentação , Grafite/química , Aptâmeros de Nucleotídeos/química , Técnicas Eletroquímicas/métodos , Desenho de Equipamento , Ferrocianetos/química , Colorimetria/instrumentação , Colorimetria/métodos
SELEÇÃO DE REFERÊNCIAS
DETALHE DA PESQUISA