Evaluating the Margins of Breast Cancer Tumors by Using Digital Breast Tomosynthesis with Deep Learning: A Preliminary Assessment.

BACKGROUND: The assessment information of tumor margins is extremely important for the success of the breast cancer surgery and whether the patient undergoes a second operation. However, conducting surgical margin assessments is a time-consuming task that requires pathology-related skills and equipment, and often cannot be provided in a timely manner. To address this challenge, digital breast tomosynthesis technology was utilized to generate detailed cross-sectional images of the breast tissue and integrate deep learning algorithms for image segmentation, achieving an assessment of tumor margins during surgery. METHODS: this study utilized post-operative tissue samples from 46 patients who underwent breast-conserving treatment, and generated image sets using digital breast tomosynthesis for the training and evaluation of deep learning models. RESULTS: Deep learning algorithms effectively identifying the tumor area. They achieved a Mean Intersection over Union (MIoU) of 0.91, global accuracy of 99%, weighted IoU of 44%, precision of 98%, recall of 83%, F1 score of 89%, and dice coefficient of 93% on the training dataset; for the testing dataset, MIoU was at 83%, global accuracy at 97%, weighted IoU at 38%, precision at 87%, recall rate at 69%, F1 score at 76%, dice coefficient at 86%. CONCLUSIONS: The initial evaluation suggests that the deep learning-based image segmentation method is highly accurate in measuring breast tumor margins. This helps provide information related to tumor margins during surgery, and by using different datasets, this research method can also be applied to the surgical margin assessment of various types of tumors.

Refined protocol for newly onset identification in non-obese diabetic mice: an animal-friendly, cost-effective, and efficient alternative.

BACKGROUND: Therapeutic interventions for diabetes are most effective when administered in the newly onset phase, yet determining the exact onset moment can be elusive in practice. Spontaneous autoimmune diabetes among NOD mice appears randomly between 12 and 32 weeks of age with an incidence range from 60 to 90%. Furthermore, the disease often progresses rapidly to severe diabetes within days, resulting in a very short window of newly onset phase, that poses significant challenge in early diagnosis. Conventionally, extensive blood glucose (BG) testing is typically required on large cohorts throughout several months to conduct prospective survey. We incorporated ultrasensitive urine glucose (UG) testing into an ordinary BG survey process, initially aiming to elucidate the lag period required for excessive glucose leaking from blood to urine during diabetes progression in the mouse model. RESULTS: The observations unexpectedly revealed that small amounts of glucose detected in the urine often coincide with, sometimes even a couple days prior than elevated BG is diagnosed. Accordingly, we conducted the UG-based survey protocol in another cohort that was validated to accurately identified every individual near onset, who could then be confirmed by following few BG tests to fulfill the consecutive BG + criteria. This approach required fewer than 95 BG tests, compared to over 700 tests with traditional BG survey, to diagnose all the 37-38 diabetic mice out of total 60. The average BG level at diagnosis was slightly below 350 mg/dl, lower than the approximately 400 mg/dl observed with conventional BG monitoring. CONCLUSIONS: We demonstrated a near perfect correlation between BG + and ultrasensitive UG + results in prospective survey with no lag period detected under twice weekly of testing frequency. This led to the refined protocol based on surveying with noninvasive UG testing, allowing for the early identification of newly onset diabetic mice with only a few BG tests required per mouse. This protocol significantly reduces the need for extensive blood sampling, lancet usage, labor, and animal distress, aligning with the 3Rs principle. It presents a convenient, accurate, and animal-friendly alternative for early diabetes diagnosis, facilitating research on diagnosis, pathogenesis, prevention, and treatment.

Improving Breast Cancer Detection and Diagnosis through Semantic Segmentation Using the Unet3+ Deep Learning Framework.

We present an analysis and evaluation of breast cancer detection and diagnosis using segmentation models. We used an advanced semantic segmentation method and a deep convolutional neural network to identify the Breast Imaging Reporting and Data System (BI-RADS) lexicon for breast ultrasound images. To improve the segmentation results, we used six models to analyse 309 patients, including 151 benign and 158 malignant tumour images. We compared the Unet3+ architecture with several other models, such as FCN, Unet, SegNet, DeeplabV3+ and pspNet. The Unet3+ model is a state-of-the-art, semantic segmentation architecture that showed optimal performance with an average accuracy of 82.53% and an average intersection over union (IU) of 52.57%. The weighted IU was found to be 89.14% with a global accuracy of 90.99%. The application of these types of segmentation models to the detection and diagnosis of breast cancer provides remarkable results. Our proposed method has the potential to provide a more accurate and objective diagnosis of breast cancer, leading to improved patient outcomes.

Mammographic Density Reduction is Associated to the Prognosis in Asian Breast Cancer Patients Receiving Hormone Therapy.

INTRODUCTION: Using mammographic density as a significant biomarker for predicting prognosis in adjuvant hormone therapy patients is controversial due to the conflicting results of recent studies. This study aimed to evaluate hormone therapy-induced mammographic density reduction and its association with prognosis in Taiwanese patients. METHODS: In this retrospective study, 1941 patients with breast cancer were screened, and 399 patients with estrogen receptor-positive breast cancer who received adjuvant hormone therapy were enrolled. The mammographic density was measured using a fully automatic estimation procedure based on full-field digital mammography. The prognosis included relapse and metastasis during treatment follow-up. The Kaplan-Meier method and Cox proportional hazards model were used for disease-free survival analysis. RESULTS: A mammographic density reduction rate >20.8%, measured preoperatively and after receiving hormone therapy from 12-18 months, was a significant threshold for predicting prognosis in patients with breast cancer. The disease-free survival rate was significantly higher in patients whose mammographic density reduction rate was >20.8% (P = .048). CONCLUSION: This study's findings could help estimate the prognosis for patients with breast cancer and may improve the quality of adjuvant hormone therapy after enlarging the study cohort in the future.

Specific Forms of Graphene Quantum Dots Induce Apoptosis and Cell Cycle Arrest in Breast Cancer Cells.

Graphene quantum dots (GQDs), nanomaterials derived from graphene and carbon dots, are highly stable, soluble, and have exceptional optical properties. Further, they have low toxicity and are excellent vehicles for carrying drugs or fluorescein dyes. Specific forms of GQDs can induce apoptosis and could be used to treat cancers. In this study, three forms of GQDs (GQD (nitrogen:carbon = 1:3), ortho-GQD, and meta-GQD) were screened and tested for their potential to inhibit breast cancer cell (MCF-7, BT-474, MDA-MB-231, and T-47D) growth. All three GQDs decreased cell viability after 72 h of treatment and specifically affected breast cancer cell proliferation. An assay for the expression of apoptotic proteins revealed that p21 and p27 were up-regulated (1.41-fold and 4.75-fold) after treatment. In particular, ortho-GQD-treated cells showed G2/M phase arrest. The GQDs specifically induced apoptosis in estrogen receptor-positive breast cancer cell lines. These results indicate that these GQDs induce apoptosis and G2/M cell cycle arrest in specific breast cancer subtypes and could potentially be used for treating breast cancers.

Semantic Segmentation of the Malignant Breast Imaging Reporting and Data System Lexicon on Breast Ultrasound Images by Using DeepLab v3.

In this study, an advanced semantic segmentation method and deep convolutional neural network was applied to identify the Breast Imaging Reporting and Data System (BI-RADS) lexicon for breast ultrasound images, thereby facilitating image interpretation and diagnosis by providing radiologists an objective second opinion. A total of 684 images (380 benign and 308 malignant tumours) from 343 patients (190 benign and 153 malignant breast tumour patients) were analysed in this study. Six malignancy-related standardised BI-RADS features were selected after analysis. The DeepLab v3+ architecture and four decode networks were used, and their semantic segmentation performance was evaluated and compared. Subsequently, DeepLab v3+ with the ResNet-50 decoder showed the best performance in semantic segmentation, with a mean accuracy and mean intersection over union (IU) of 44.04% and 34.92%, respectively. The weighted IU was 84.36%. For the diagnostic performance, the area under the curve was 83.32%. This study aimed to automate identification of the malignant BI-RADS lexicon on breast ultrasound images to facilitate diagnosis and improve its quality. The evaluation showed that DeepLab v3+ with the ResNet-50 decoder was suitable for solving this problem, offering a better balance of performance and computational resource usage than a fully connected network and other decoders.

PBK Expression Is Associated With Prognosis of Patients With Oral Squamous Cell Carcinoma Treated With Radiotherapy: A Retrospective Study.

BACKGROUND/AIM: To investigate the impact of PDZ-binding kinase (PBK) on the clinical outcome of patients with oral squamous cell carcinoma (OSCC) who received radiotherapy. PATIENTS AND METHODS: PBK immunoreactivity of cancer specimens obtained from 179 patients with primary OSCC was analyzed by immunohistochemistry. RESULTS: High PBK expression in tumor cells tended to be associated with advanced N-stage. The 5-year survival rate was greater for patients with high total PBK expression than in those with low PBK expression. After adjustment, high PBK remained associated with a favorable outcome. In subgroups according to tumor stage, the prognostic role was significant in patients with stage III/IV rather than those with stage I/II disease. CONCLUSION: We suggest that PBK expression should be used as an independent prognostic marker for patients with OSCC treated with radiotherapy, especially for those with advanced-stage disease.

Classification of malignant tumours in breast ultrasound using unsupervised machine learning approaches.

Traditional computer-aided diagnosis (CAD) processes include feature extraction, selection, and classification. Effective feature extraction in CAD is important in improving the classification's performance. We introduce a machine-learning method and have designed an analysis procedure of benign and malignant breast tumour classification in ultrasound (US) images without a need for a priori tumour region-selection processing, thereby decreasing clinical diagnosis efforts while maintaining high classification performance. Our dataset constituted 677 US images (benign: 312, malignant: 365). Regarding two-dimensional US images, the oriented gradient descriptors' histogram pyramid was extracted and utilised to obtain feature vectors. The correlation-based feature selection method was used to evaluate and select significant feature sets for further classification. Sequential minimal optimisation-combining local weight learning-was utilised for classification and performance enhancement. The image dataset's classification performance showed an 81.64% sensitivity and 87.76% specificity for malignant images (area under the curve = 0.847). The positive and negative predictive values were 84.1 and 85.8%, respectively. Here, a new workflow, utilising machine learning to recognise malignant US images was proposed. Comparison of physician diagnoses and the automatic classifications made using machine learning yielded similar outcomes. This indicates the potential applicability of machine learning in clinical diagnoses.

Incorporating the Breast Imaging Reporting and Data System Lexicon with a Fully Convolutional Network for Malignancy Detection on Breast Ultrasound.

In this study, we applied semantic segmentation using a fully convolutional deep learning network to identify characteristics of the Breast Imaging Reporting and Data System (BI-RADS) lexicon from breast ultrasound images to facilitate clinical malignancy tumor classification. Among 378 images (204 benign and 174 malignant images) from 189 patients (102 benign breast tumor patients and 87 malignant patients), we identified seven malignant characteristics related to the BI-RADS lexicon in breast ultrasound. The mean accuracy and mean IU of the semantic segmentation were 32.82% and 28.88, respectively. The weighted intersection over union was 85.35%, and the area under the curve was 89.47%, showing better performance than similar semantic segmentation networks, SegNet and U-Net, in the same dataset. Our results suggest that the utilization of a deep learning network in combination with the BI-RADS lexicon can be an important supplemental tool when using ultrasound to diagnose breast malignancy.

Classification of malignant tumors in breast ultrasound using a pretrained deep residual network model and support vector machine.

In this study, a transfer learning method was utilized to recognize and classify benign and malignant breast tumors, using two-dimensional breast ultrasound (US) images, to decrease the effort expended by physicians and improve the quality of clinical diagnosis. The pretrained deep residual network model was utilized for image feature extraction from the convolutional layer of the trained network; whereas, the linear support vector machine (SVM), with a sequential minimal optimization solver, was used to classify the extracted feature. We used an image dataset with 2099 unlabeled two-dimensional breast US images, collected from 543 patients (benign: 302, malignant: 241). The classification performance yielded a sensitivity of 94.34 % and a specificity of 93.22 % for malignant images (Area under curve = 0.938). The positive and negative predictive values were 92.6 and 94.8, respectively. A comparison between the diagnosis made by the physician and the automated classification by a trained classifier, showed that the latter had significantly better outcomes. This indicates the potential applicability of the proposed approach that incorporates both the pretrained deep learning network and a well-trained classifier, to improve the quality and efficacy of clinical diagnosis.

Multiview Contouring for Breast Tumor on Magnetic Resonance Imaging.

The shape and contour of the lesion are shown to be effective features for physicians to identify breast tumor as benign or malignant. The region of the lesion is usually manually created by the physician according to their clinical experience; therefore, contouring tumors on breast magnetic resonance imaging (MRI) is difficult and time-consuming. For this purpose, an automatic contouring method for breast tumors was developed for less burden in the analysis and to decrease the observed bias to help in making decisions clinically. In this study, a multiview segmentation method for detecting and contouring breast tumors in MRI was represented. The preprocessing of the proposed method reduces any amount of noises but preserves the shape and contrast of the breast tumor. The two-dimensional (2D) level-set segmentation method extracts contours of breast tumors from the transverse, coronal, and sagittal planes. The obtained contours are further utilized to generate appropriate three-dimensional (3D) contours. Twenty breast tumor cases were evaluated and the simulation results show that the proposed contouring method was an efficient method for delineating 3D contours of breast tumors in MRI.

Implant volume estimation in direct-to-implant breast reconstruction after nipple-sparing mastectomy.

BACKGROUND: Nipple-sparing mastectomy (NSM) is an increasingly popular alternative to more traditional mastectomy approaches. However, estimating the implant volume during direct-to-implant (DTI) reconstruction following NSM is difficult for surgeons with little-to-moderate experience. We aimed to provide a fast, easy to use, and accurate method to aid in the estimation of implant size for DTI reconstruction using the specimen weight and breast volume. METHODS: A retrospective analysis was performed using data from 145 NSM patients with specific implant types. Standard two-dimensional digital mammograms were obtained in 118 of the patients. Breast morphological factors (specimen weight, mammographic breast density and volume, and implant size and type) were recorded. Curve-fitting and linear regression models were used to develop formulas predicting the implant volume, and the prediction performance of the obtained formulas was evaluated using the prospective data set. RESULTS: Two formulas to estimate the implant size were obtained, one using the specimen weight and one using the breast volume. The coefficients of correlation (R2) in these formulas were over 0.98 and the root mean squared errors were approximately 13. CONCLUSIONS: These implant volume estimate formulas benefit surgeons by providing a preoperative implant volume assessment in DTI reconstruction using the breast volume and an intraoperative assessment using the specimen weight. The implant size estimation formulas obtained in the present study may be applied in a majority of patients.

Mammographic Density Distribution of Healthy Taiwanese Women and its Naturally Decreasing Trend with Age.

We analysed typical mammographic density (MD) distributions of healthy Taiwanese women to augment existing knowledge, clarify cancer risks, and focus public health efforts. From January 2011 to December 2015, 88,193 digital mammograms were obtained from 69,330 healthy Taiwanese women (average, 1.27 mammograms each). MD measurements included dense volume (DV) and volumetric density percentage (VPD) and were quantified by fully automated volumetric density estimation and Box-Cox normalization. Prediction of the declining MD trend was estimated using curve fitting and a rational model. Normalized DV and VPD Lowess curves demonstrated similar but non-identical distributions. In high-density grade participants, the VPD increased from 12.45% in the 35-39-year group to 13.29% in the 65-69-year group but only from 5.21% to 8.47% in low-density participants. Regarding the decreased cumulative VPD percentage, the mean MD declined from 12.79% to 19.31% in the 45-50-year group versus the 50-55-year group. The large MD decrease in the fifth decade in this present study was similar to previous observations of Western women. Obtaining an MD distribution model with age improves the understanding of breast density trends and age variations and provides a reference for future studies on associations between MD and cancer risk.

Using Flow Characteristics in Three-Dimensional Power Doppler Ultrasound Imaging to Predict Complete Responses in Patients Undergoing Neoadjuvant Chemotherapy.

OBJECTIVES: Strategies are needed for the identification of a poor response to treatment and determination of appropriate chemotherapy strategies for patients in the early stages of neoadjuvant chemotherapy for breast cancer. We hypothesize that power Doppler ultrasound imaging can provide useful information on predicting response to neoadjuvant chemotherapy. METHODS: The solid directional flow of vessels in breast tumors was used as a marker of pathologic complete responses (pCR) in patients undergoing neoadjuvant chemotherapy. Thirty-one breast cancer patients who received neoadjuvant chemotherapy and had tumors of 2 to 5 cm were recruited. Three-dimensional power Doppler ultrasound with high-definition flow imaging technology was used to acquire the indices of tumor blood flow/volume, and the chemotherapy response prediction was established, followed by support vector machine classification. RESULTS: The accuracy of pCR prediction before the first chemotherapy treatment was 83.87% (area under the ROC curve [AUC] = 0.6957). After the second chemotherapy treatment, the accuracy of was 87.9% (AUC = 0.756). Trend analysis showed that good and poor responders exhibited different trends in vascular flow during chemotherapy. CONCLUSIONS: This preliminary study demonstrates the feasibility of using the vascular flow in breast tumors to predict chemotherapeutic efficacy.

Effectiveness of evaluating tumor vascularization using 3D power Doppler ultrasound with high-definition flow technology in the prediction of the response to neoadjuvant chemotherapy for T2 breast cancer: a preliminary report.

The aim of this study was to evaluate the effectiveness of advanced ultrasound (US) imaging of vascular flow and morphological features in the prediction of a pathologic complete response (pCR) and a partial response (PR) to neoadjuvant chemotherapy for T2 breast cancer.Twenty-nine consecutive patients with T2 breast cancer treated with six courses of anthracycline-based neoadjuvant chemotherapy were enrolled. Three-dimensional (3D) power Doppler US with high-definition flow (HDF) technology was used to investigate the blood flow in and morphological features of the tumors. Six vascularity quantization features, three morphological features, and two vascular direction features were selected and extracted from the US images. A support vector machine was used to evaluate the changes in vascularity after neoadjuvant chemotherapy, and pCR and PR were predicted on the basis of these changes.The most accurate prediction of pCR was achieved after the first chemotherapy cycle, with an accuracy of 93.1% and a specificity of 85.5%, while that of a PR was achieved after the second cycle, with an accuracy of 79.31% and a specificity of 72.22%.Vascularity data can be useful to predict the effects of neoadjuvant chemotherapy. Determination of changes in vascularity after neoadjuvant chemotherapy using 3D power Doppler US with HDF can generate accurate predictions of the patient response, facilitating early decision-making.

Whole-genome sequencing and identification of Morganella morganii KT pathogenicity-related genes.

BACKGROUND: The opportunistic enterobacterium, Morganella morganii, which can cause bacteraemia, is the ninth most prevalent cause of clinical infections in patients at Changhua Christian Hospital, Taiwan. The KT strain of M. morganii was isolated during postoperative care of a cancer patient with a gallbladder stone who developed sepsis caused by bacteraemia. M. morganii is sometimes encountered in nosocomial settings and has been causally linked to catheter-associated bacteriuria, complex infections of the urinary and/or hepatobiliary tracts, wound infection, and septicaemia. M. morganii infection is associated with a high mortality rate, although most patients respond well to appropriate antibiotic therapy. To obtain insights into the genome biology of M. morganii and the mechanisms underlying its pathogenicity, we used Illumina technology to sequence the genome of the KT strain and compared its sequence with the genome sequences of related bacteria. RESULTS: The 3,826,919-bp sequence contained in 58 contigs has a GC content of 51.15% and includes 3,565 protein-coding sequences, 72 tRNA genes, and 10 rRNA genes. The pathogenicity-related genes encode determinants of drug resistance, fimbrial adhesins, an IgA protease, haemolysins, ureases, and insecticidal and apoptotic toxins as well as proteins found in flagellae, the iron acquisition system, a type-3 secretion system (T3SS), and several two-component systems. Comparison with 14 genome sequences from other members of Enterobacteriaceae revealed different degrees of similarity to several systems found in M. morganii. The most striking similarities were found in the IS4 family of transposases, insecticidal toxins, T3SS components, and proteins required for ethanolamine use (eut operon) and cobalamin (vitamin B12) biosynthesis. The eut operon and the gene cluster for cobalamin biosynthesis are not present in the other Proteeae genomes analysed. Moreover, organisation of the 19 genes of the eut operon differs from that found in the other non-Proteeae enterobacterial genomes. CONCLUSIONS: This is the first genome sequence of M. morganii, which is a clinically relevant pathogen. Comparative genome analysis revealed several pathogenicity-related genes and novel genes not found in the genomes of other members of Proteeae. Thus, the genome sequence of M. morganii provides important information concerning virulence and determinants of fitness in this pathogen.

Genetic copy number variants in myocardial infarction patients with hyperlipidemia.

BACKGROUND: Cardiovascular disease is the chief cause of death in Taiwan and many countries, of which myocardial infarction (MI) is the most serious condition. Hyperlipidemia appears to be a significant cause of myocardial infarction, because it causes atherosclerosis directly. In recent years, copy number variation (CNV) has been analyzed in genomewide association studies of complex diseases. In this study, CNV was analyzed in blood samples and SNP arrays from 31 myocardial infarction patients with hyperlipidemia. RESULTS: We identified seven CNV regions that were associated significantly with hyperlipidemia and myocardial infarction in our patients through multistage analysis (P<0.001), at 1p21.3, 1q31.2 (CDC73), 1q42.2 (DISC1), 3p21.31 (CDCP1), 10q11.21 (RET) 12p12.3 (PIK3C2G) and 16q23.3 (CDH13), respectively. In particular, the CNV region at 10q11.21 was examined by quantitative real-time PCR, the results of which were consistent with microarray findings. CONCLUSIONS: Our preliminary results constitute an alternative method of evaluating the relationship between CNV regions and cardiovascular disease. These susceptibility CNV regions may be used as biomarkers for early-stage diagnosis of hyperlipidemia and myocardial infarction, rendering them valuable for further research and discussion.