- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Automated diagnosis of breast ultrasonography images using deep neural networks
摘要： Ultrasonography images of breast mass aid in the detection and diagnosis of breast cancer. Manually analyzing ultrasonography images is time-consuming, exhausting and subjective. Automated analyzing such images is desired. In this study, we develop an automated breast cancer diagnosis model for ultra-sonography images. Traditional methods of automated ultrasonography images analysis employ hand-crafted features to classify images, and lack robustness to the variation in the shapes, size and texture of breast lesions, leading to low sensitivity in clinical applications. To overcome these shortcomings, we propose a method to diagnose breast ultrasonography images using deep convolutional neural networks with multi-scale kernels and skip connections. Our method consists of two components: the first one is to determine whether there are ma-lignant tumors in the image, and the second one is to recognize solid nodules. In order to let the two networks work in a collaborative way, a region enhance mechanism based on class activation maps is proposed. The mechanism helps to improve classification accuracy and sensitivity for both networks. A cross training algorithm is introduced to train the networks. We construct a large annotated dataset containing a total of 8145 breast ultrasonography images to train and evaluate the models. All of the annotations are proven by patho-logical records. The proposed method is compared with two state-of-the-art approaches, and outperforms both of them by a large margin. Experimental results show that our approach achieves a performance comparable to human sonographers and can be applied to clinical scenarios.
关键词： ultrasonography,breast cancer,deep neural networks
更新于2025-09-23 15:19:57
-
Letter to the Editor: A response to Hruska’s case study on molecular breast imaging and the need for true tissue quantification
摘要： We applaud the efforts by Hruska et al. to quantify differences in tissue using molecular breast imaging (MBI) and background parenchymal uptake (BPU); we have discussed the use of such previously [2]. The approach while commendable did not provide diagnostically useful information to differentiate tissue types. This approach, like the utilization of standardized uptake value (SUV), compares differences in background with tissue [3]. As we have already discussed [2, 4, 5] in the literature, this approach is an incorrect model, due to (1) the critical lack of standardization and calibration of nuclear cameras including both SPECT/Planar and PET; (2) the utilization of ratios which are not absolute values and therefore cannot be used to differentiate tissue based upon those issues, critical to the understanding of tissue differences; and (3) the inability to truly 'measure' transitional changes in tissue, which would allow for the determination of actual treatment response on a per patient basis, saving time, money, and lives.
关键词： Patent protected,AI,Breast cancer,FMTVDM??,Theranostics,Quantification,B.E.S.T. Imaging??,Nuclear camera quantitative calibration,Breast inflammation
更新于2025-09-19 17:15:36
-
Determination of the Concentration of Polyamines with SPR-based Immune Biosensor for Early Diagnostics of Breast Cancer
摘要： The paper presents the results of research on the development of immune biosensor test system for express detection of polyamines in cells of breast cancer. Determination of polyamines was performed by using an analytical device - immune biosensor based on surface plasmon resonance (SPR), where "antigen-antibody" reaction is performed in real time on the surface of transducer, resulting in formation of immune complexes and recording the shift of resonance angle. The method can detect the studied polyamines spermine and spermidine in concentration less than 10 ng in 1 mL, and with increasing concentrations a statistical probability of the analysis' result rise sharply. Moreover, the dependence of sensitivity of biosensor response from the concentration of polyamines is in the range of 10-100 ng /mL.
关键词： Antigen,Antibody,Breast cancer,Surface plasmon resonance,Polyamines
更新于2025-09-19 17:15:36
-
Features of Undiagnosed Breast Cancers at Screening Breast MR Imaging and Potential Utility of Computer-Aided Evaluation
摘要： Objective: To retrospectively evaluate the features of undiagnosed breast cancers on prior screening breast magnetic resonance (MR) images in patients who were subsequently diagnosed with breast cancer, as well as the potential utility of MR-computer-aided evaluation (CAE). Materials and Methods: Between March 2004 and May 2013, of the 72 consecutive pairs of prior negative MR images and subsequent MR images with diagnosed cancers (median interval, 32.8 months; range, 5.4–104.6 months), 36 (50%) had visible findings (mean size, 1.0 cm; range, 0.3–5.2 cm). The visible findings were divided into either actionable or underthreshold groups by the blinded review by 5 radiologists. MR imaging features, reasons for missed cancer, and MR-CAE features according to actionability were evaluated. Results: Of the 36 visible findings on prior MR images, 33.3% (12 of 36) of the lesions were determined to be actionable and 66.7% (24 of 36) were underthreshold; 85.7% (6 of 7) of masses and 31.6% (6 of 19) of non-mass enhancements were classified as actionable lesions. Mimicking physiologic enhancements (27.8%, 10 of 36) and small lesion size (27.8%, 10 of 36) were the most common reasons for missed cancer. Actionable findings tended to show more washout or plateau kinetic patterns on MR-CAE than underthreshold findings, as the 100% of actionable findings and 46.7% of underthreshold findings showed washout or plateau (p = 0.008). Conclusion: MR-CAE has the potential for reducing the number of undiagnosed breast cancers on screening breast MR images, the majority of which are caused by mimicking physiologic enhancements or small lesion size.
关键词： False negative breast cancer,Computer-aided evaluation,Magnetic resonance imaging
更新于2025-09-19 17:15:36
-
An Oxygen Self-sufficient Fluorinated Nanoplatform for Relieved Tumor Hypoxia and Enhanced Photodynamic Therapy of Cancers
摘要： The efficacy of photodynamic therapy (PDT) in the solid tumor is hampered by many challenges, including its oxygen self-consuming nature, insufficient oxygen levels within the hypoxic tumor microenvironment, and limited penetration of photosensitizers within tumors. Herein, we develop the IR780@O2-SFNs/iRGD as an oxygen self-sufficient and tumor-penetrating nanoplatform, which consists of IR780 loaded pH-sensitive fluorocarbon functionalized nanoparticles (SFNs) and iRGD as a tumor targeting peptide that can penetrate deeper within the tumor. Because of the high oxygen affinity and outstanding permeability of the obtained nanoplatform, oxygen and IR780 which are encapsulated in the same core can play their roles to the utmost, resulting in remarkably accelerated singlet oxygen production, as demonstrated in vitro by the 3D multicellular spheroids and in vivo by tumor tissues. More interestingly, a single-dose intravenous administration of IR780@O2-SFNs/iRGD into mice bearing orthotopic breast cancer could selectively accumulate at the tumor site, highly alleviate the tumor hypoxia, significantly inhibit the primary tumor growth and reduce the lung and liver metastasis, enabling the improved photodynamic therapeutic performance. Thus, this work paves an effective way to improve PDT efficacy through increasing tumor oxygenation and selective delivery of photosensitizers to the deep and hypoxic tumor.
关键词： oxygen self-sufficient,tumor penetration,tumor oxygenation,photodynamic therapy,orthotopic breast cancer
更新于2025-09-19 17:15:36
-
Early Therapeutic Prediction Based on Tumor Hemodynamic Response Imaging: Clinical Studies in Breast Cancer with Time-Resolved Diffuse Optical Spectroscopy
摘要： This study reports data from three clinical studies using the time-resolved diffuse optical spectroscopy (TRS) system among breast cancer patients. The parameters of oxy-hemoglobin (O2Hb), deoxy-hemoglobin (HHb), total hemoglobin (tHb), and oxygen saturation (SO2) were evaluated using TRS, and its efficacy was tested in three trials. In trial 1, we recruited 118 patients with primary breast cancer to estimate the tumor detection rate. The cumulative detection rate was 62.7%, while that in T stage 0 was 31.3% and in T stage 1 was 44.7%. These were lower than those of T stage 2 (78.9%) and T stage 3 (100%). Next, we used TRS to monitor tumor hemodynamic response to neoadjuvant chemotherapy (n = 100) and found that pathological complete response (pCR) tumors had significantly lower tumor tHb than non-pCR tumors; a similar result was observed in estrogen receptor (ER)-negative tumors, but not in ER-positive tumors. The third trial monitored hemodynamic response to antiangiogenic therapy, bevacizumab (n = 28), and we demonstrated that sequential optical measurement of tumor SO2 might be useful for detecting acute hypoxia 1–3 days after bevacizumab initiation. Next, response monitoring of neoadjuvant endocrine therapy (n = 30) suggested that changes in tumor tHb during treatment can predict and distinguish between responsive and non-responsive tumors early in letrozole therapy. In conclusion, our results show that hemodynamic monitoring of tumors by TRS could pair the unique features of tumor physiology to drug therapy and contribute to patient-tailored medicine. We recently established a platform for performing TRS in patients with breast cancer.
关键词： diffuse optical spectroscopy,chemotherapy,breast cancer
更新于2025-09-19 17:15:36
-
Multiparametric monitoring of chemotherapy treatment response in locally advanced breast cancer using quantitative ultrasound and diffuse optical spectroscopy
摘要： Purpose: This study evaluated pathological response to neoadjuvant chemotherapy using quantitative ultrasound (QUS) and diffuse optical spectroscopy imaging (DOSI) biomarkers in locally advanced breast cancer (LABC). Materials and Methods: The institution’s ethics review board approved this study. Subjects (n = 22) gave written informed consent prior to participating. US and DOSI data were acquired, relative to the start of neoadjuvant chemotherapy, at weeks 0, 1, 4, 8 and preoperatively. QUS parameters including the mid-band fit (MBF), 0-MHz intercept (SI), and the spectral slope (SS) were determined from tumor ultrasound data using spectral analysis. In the same patients, DOSI was used to measure parameters relating to tumor hemoglobin and composition. Discriminant analysis and receiver-operating characteristic (ROC) analysis was used to classify clinical and pathological response during treatment and to estimate the area under the curve (AUC). Additionally, multivariate analysis was carried out for pairwise QUS/DOSI parameter combinations using a logistic regression model. Results: Individual QUS and DOSI parameters, including the (SI), oxy-hemoglobin (HbO2), and total hemoglobin (HbT) were significant markers for response after one week of treatment (p < 0.01). Multivariate (pairwise) combinations increased the sensitivity, specificity and AUC at this time; the SI + HbO2 showed a sensitivity/specificity of 100%, and an AUC of 1.0. Conclusions: QUS and DOSI demonstrated potential as coincident markers for treatment response and may potentially facilitate response-guided therapies. Multivariate QUS and DOSI parameters increased the sensitivity and specificity of classifying LABC patients as early as one week after treatment.
关键词： diffuse optical spectroscopy,quantitative ultrasound,treatment monitoring,neoadjuvant chemotherapy,locally advanced breast cancer
更新于2025-09-19 17:15:36
-
Anti-cancer effect of gallic acid in presence of low level laser irradiation: ROS production and induction of apoptosis and ferroptosis
摘要： Background: There are different treatments for breast cancer and melanoma that mostly have some side effects. One of the therapeutic strategies is the use of natural components. Phenol components are a class of antioxidants in plants that have many biological functions like anticancer effects. Gallic acid (GA) is a natural polyhydroxy phenolic compound and commonly found in various foods. In the present study, GA effects alone and in combination with low-level laser irradiation on human dermal fibroblast cell line (HDF), human non-tumorigenic breast epithelial cell line (MCF10A), breast cancer cell line (MDA-MB-231) and melanoma cancer cell line (A375) was under the investigation. Methods: The normal and cancerous cell lines were exposed to 660 nm low-level laser with 3 J/cm2 for 90 s. Then, the cells were treated with different concentrations of GA for 24 h. In another study, the cell lines firstly were treated with GA and then exposed to low-level laser irradiation. The effects of GA and low-level laser on cell survival and apoptosis were examined using MTT assay, light microscopy, ROS production assay, fluorescence microscopy (AO/EB double staining) and flow cytometry. Results: The results showed that pre-treatment with low-level laser and then GA reduced the survival of breast cancer cells and melanoma more than the first treatment with GA and then low-level laser irradiation. Our findings showed that ROS production in cells treated with both low-level laser and GA was more than the cells treated with GA alone. The apoptosis and ferroptosis assays confirmed the MTT results which combination treatment with low-level laser and then GA increase the cell death probably via apoptosis and ferroptosis cell death mechanisms compared to GA alone. Conclusions: This study suggests that low-level laser irradiation alone is not able to cause death in human normal and cancerous cells. Preirradiation followed by GA treatment did not change the cell viability in human normal significantly but reduces the cell survival of cancer cells more than GA alone.
关键词： Breast cancer,Apoptosis,Low level laser irradiation,Ferroptosis,Melanoma cancer,Gallic acid
更新于2025-09-19 17:13:59
-
Her2-Functionalized Gold-Nanoshelled Magnetic Hybrid Nanoparticles: a Theranostic Agent for Dual-Modal Imaging and Photothermal Therapy of Breast Cancer
摘要： Targeted theranostic platform that integrates multi-modal imaging and therapeutic function is emerging as a promising strategy for earlier detection and precise treatment of cancer. Herein, we designed targeted gold-nanoshelled poly (lactic-co-glycolic acid) (PLGA) magnetic hybrid nanoparticles carrying anti-human epidermal growth factor receptor 2 (Her2) antibodies (Her2-GPH NPs) for dual-modal ultrasound (US)/magnetic resonance (MR) imaging and photothermal therapy of breast cancer. The agent was fabricated by coating gold nanoshell around PLGA nanoparticles co-loaded with perfluorooctyl bromide (PFOB) and superparamagnetic iron oxide nanoparticles (SPIOs), followed by conjugating with anti-Her2 antibodies. Cell-targeting studies demonstrated receptor-mediated specific binding of the agent to Her2-positive human breast cancer SKBR3 cells, and its binding rate was significantly higher than that of Her2-negative cells (P < 0.001). In vitro, the agent had capabilities for contrast-enhanced US imaging as well as T2-weighted MR imaging with a relatively high relaxivity (r2 = 441.47 mM?1 s?1). Furthermore, the Her2 functionalization of the agent prominently enhanced the US/MR molecular imaging effect of targeted cells by cell-specific binding. Live/dead cell assay and targeted photothermal cytotoxicity experiments confirmed that Her2-GPH NPs could serve as effective photoabsorbers to specifically induce SKBR3 cell death upon near-infrared laser irradiation. In summary, Her2-GPH NPs were demonstrated to be novel targeted theranostic agents with great potential to facilitate early non-invasive diagnosis and adjuvant therapy of breast cancer.
关键词： Anti-Her2 antibody,Photothermal therapy,Ultrasound imaging,Magnetic resonance imaging,Theranostic agent,Breast cancer
更新于2025-09-16 10:30:52
-
Laser cleavable probes for <i>in situ</i> multiplexed glycan detection by single cell mass spectrometry
摘要： Glycans binding on the cell surface through glycosylation play a key role in controlling various cellular processes, and glycan analysis at a single-cell level is necessary to study cellular heterogeneity and diagnose diseases in the early stage. Herein, we synthesized a series of laser cleavable probes, which could sensitively detect glycans on single cells and tissues by laser desorption ionization mass spectrometry (LDI-MS). This multiplexed and quantitative glycan detection was applied to evaluate the alterations of four types of glycans on breast cancer cells and drug-resistant cancer cells at a single-cell level, indicating that drug resistance may be related to the upregulation of glycan with a b-D-galactoside (Galb) group and Neu5Aca2-6Gal(NAc)-R. Moreover, the glycan spatial distribution in cancerous and paracancerous human tissues was also demonstrated by MS imaging, showing that glycans are overexpressed in cancerous tissues. Therefore, this single-cell MS approach exhibits promise for application in studying glycan functions which are essential for clinical biomarker discovery and diagnosis of related diseases.
关键词： mass spectrometry,breast cancer,drug resistance,laser cleavable probes,single-cell analysis,glycans
更新于2025-09-16 10:30:52