The molecular mechanisms driving Plasmodium cell division.

Malaria, a vector borne disease, is a major global health and socioeconomic problem caused by the apicomplexan protozoan parasite Plasmodium. The parasite alternates between mosquito vector and vertebrate host, with meiosis in the mosquito and proliferative mitotic cell division in both hosts. In the canonical eukaryotic model, cell division is either by open or closed mitosis and karyokinesis is followed by cytokinesis; whereas in Plasmodium closed mitosis is not directly accompanied by concomitant cell division. Key molecular players and regulatory mechanisms of this process have been identified, but the pivotal role of certain protein complexes and the post-translational modifications that modulate their actions are still to be deciphered. Here, we discuss recent evidence for the function of known proteins in Plasmodium cell division and processes that are potential novel targets for therapeutic intervention. We also identify key questions to open new and exciting research to understand divergent Plasmodium cell division.

DF-dRVFL: A novel deep feature based classifier for breast mass classification.

Amongst all types of cancer, breast cancer has become one of the most common cancers in the UK threatening millions of people's health. Early detection of breast cancer plays a key role in timely treatment for morbidity reduction. Compared to biopsy, which takes tissues from the lesion for further analysis, image-based methods are less time-consuming and pain-free though they are hampered by lower accuracy due to high false positivity rates. Nevertheless, mammography has become a standard screening method due to its high efficiency and low cost with promising performance. Breast mass, as the most palpable symptom of breast cancer, has received wide attention from the community. As a result, the past decades have witnessed the speeding development of computer-aided systems that are aimed at providing radiologists with useful tools for breast mass analysis based on mammograms. However, the main issues of these systems include low accuracy and require enough computational power on a large scale of datasets. To solve these issues, we developed a novel breast mass classification system called DF-dRVFL. On the public dataset DDSM with more than 3500 images, our best model based on deep random vector functional link network showed promising results through five-cross validation with an averaged AUC of 0.93 and an average accuracy of 81.71%. Compared to sole deep learning based methods, average accuracy has increased by 0.38. Compared with the state-of-the-art methods, our method showed better performance considering the number of images for evaluation and the overall accuracy.

Plasmodium ARK2 and EB1 drive unconventional spindle dynamics, during chromosome segregation in sexual transmission stages.

The Aurora family of kinases orchestrates chromosome segregation and cytokinesis during cell division, with precise spatiotemporal regulation of its catalytic activities by distinct protein scaffolds. Plasmodium spp., the causative agents of malaria, are unicellular eukaryotes with three unique and highly divergent aurora-related kinases (ARK1-3) that are essential for asexual cellular proliferation but lack most canonical scaffolds/activators. Here we investigate the role of ARK2 during sexual proliferation of the rodent malaria Plasmodium berghei, using a combination of super-resolution microscopy, mass spectrometry, and live-cell fluorescence imaging. We find that ARK2 is primarily located at spindle microtubules in the vicinity of kinetochores during both mitosis and meiosis. Interactomic and co-localisation studies reveal several putative ARK2-associated interactors including the microtubule-interacting protein EB1, together with MISFIT and Myosin-K, but no conserved eukaryotic scaffold proteins. Gene function studies indicate that ARK2 and EB1 are complementary in driving endomitotic division and thereby parasite transmission through the mosquito. This discovery underlines the flexibility of molecular networks to rewire and drive unconventional mechanisms of chromosome segregation in the malaria parasite.

Meiosis in Plasmodium: how does it work?

Meiosis is sexual cell division, a process in eukaryotes whereby haploid gametes are produced. Compared to canonical model eukaryotes, meiosis in apicomplexan parasites appears to diverge from the process with respect to the molecular mechanisms involved; the biology of Plasmodium meiosis, and its regulation by means of post-translational modification, are largely unexplored. Here, we discuss the impact of technological advances in cell biology, evolutionary bioinformatics, and genome-wide functional studies on our understanding of meiosis in the Apicomplexa. These parasites, including Plasmodium falciparum, Toxoplasma gondii, and Eimeria spp., have significant socioeconomic impact on human and animal health. Understanding this key stage during the parasite's life cycle may well reveal attractive targets for therapeutic intervention.

Plasmodium ARK2-EB1 axis drives the unconventional spindle dynamics, scaffold formation and chromosome segregation of sexual transmission stages.

Mechanisms of cell division are remarkably diverse, suggesting the underlying molecular networks among eukaryotes differ extensively. The Aurora family of kinases orchestrates the process of chromosome segregation and cytokinesis during cell division through precise spatiotemporal regulation of their catalytic activities by distinct scaffolds. Plasmodium spp., the causative agents of malaria, are unicellular eukaryotes that have three divergent aurora-related kinases (ARKs) and lack most canonical scaffolds/activators. The parasite uses unconventional modes of chromosome segregation during endomitosis and meiosis in sexual transmission stages within mosquito host. This includes a rapid threefold genome replication from 1N to 8N with successive cycles of closed mitosis, spindle formation and chromosome segregation within eight minutes (termed male gametogony). Kinome studies had previously suggested likely essential functions for all three Plasmodium ARKs during asexual mitotic cycles; however, little is known about their location, function, or their scaffolding molecules during unconventional sexual proliferative stages. Using a combination of super-resolution microscopy, mass spectrometry, and live-cell fluorescence imaging, we set out to investigate the role of the atypical Aurora paralog ARK2 to proliferative sexual stages using rodent malaria model Plasmodium berghei . We find that ARK2 primarily localises to the spindle apparatus in the vicinity of kinetochores during both mitosis and meiosis. Interactomics and co-localisation studies reveal a unique ARK2 scaffold at the spindle including the microtubule plus end-binding protein EB1, lacking conserved Aurora scaffold proteins. Gene function studies indicate complementary functions of ARK2 and EB1 in driving endomitotic divisions and thereby parasite transmission. Our discovery of a novel Aurora kinase spindle scaffold underlines the emerging flexibility of molecular networks to rewire and drive unconventional mechanisms of chromosome segregation in the malaria parasite Plasmodium .

Plasmodium ARK2-EB1 axis drives the unconventional spindle dynamics, scaffold formation and chromosome segregation of sexual transmission stages.

Mechanisms of cell division are remarkably diverse, suggesting the underlying molecular networks among eukaryotes differ extensively. The Aurora family of kinases orchestrates the process of chromosome segregation and cytokinesis during cell division through precise spatiotemporal regulation of their catalytic activities by distinct scaffolds. Plasmodium spp., the causative agents of malaria, are unicellular eukaryotes that have three divergent aurora-related kinases (ARKs) and lack most canonical scaffolds/activators. The parasite uses unconventional modes of chromosome segregation during endomitosis and meiosis in sexual transmission stages within mosquito host. This includes a rapid threefold genome replication from 1N to 8N with successive cycles of closed mitosis, spindle formation and chromosome segregation within eight minutes (termed male gametogony). Kinome studies had previously suggested likely essential functions for all three Plasmodium ARKs during asexual mitotic cycles; however, little is known about their location, function, or their scaffolding molecules during unconventional sexual proliferative stages. Using a combination of super-resolution microscopy, mass spectrometry, omics and live-cell fluorescence imaging, we set out to investigate the contribution of the atypical Aurora paralog ARK2 to proliferative sexual stages using rodent malaria model Plasmodium berghei. We find that ARK2 primarily localises to the spindle apparatus associated with kinetochores during both mitosis and meiosis. Interactomics and co-localisation studies reveal a unique ARK2 scaffold at the spindle including the microtubule plus end-binding protein EB1 and lacking some other conserved molecules. Gene function studies indicate complementary functions of ARK2 and EB1 in driving endomitotic divisions and thereby parasite transmission. Our discovery of a novel Aurora spindle scaffold underlines the emerging flexibility of molecular networks to rewire and drive unconventional mechanisms of chromosome segregation in the malaria parasite Plasmodium.

Genome engineering for estrogen receptor mutations reveals differential responses to anti-estrogens and new prognostic gene signatures for breast cancer.

Mutations in the estrogen receptor (ESR1) gene are common in ER-positive breast cancer patients who progress on endocrine therapies. Most mutations localise to just three residues at, or near, the C-terminal helix 12 of the hormone binding domain, at leucine-536, tyrosine-537 and aspartate-538. To investigate these mutations, we have used CRISPR-Cas9 mediated genome engineering to generate a comprehensive set of isogenic mutant breast cancer cell lines. Our results confirm that L536R, Y537C, Y537N, Y537S and D538G mutations confer estrogen-independent growth in breast cancer cells. Growth assays show mutation-specific reductions in sensitivities to drugs representing three classes of clinical anti-estrogens. These differential mutation- and drug-selectivity profiles have implications for treatment choices following clinical emergence of ER mutations. Our results further suggest that mutant expression levels may be determinants of the degree of resistance to some anti-estrogens. Differential gene expression analysis demonstrates up-regulation of estrogen-responsive genes, as expected, but also reveals that enrichment for interferon-regulated gene expression is a common feature of all mutations. Finally, a new gene signature developed from the gene expression profiles in ER mutant cells predicts clinical response in breast cancer patients with ER mutations.

Shallow WGS of individual CTCs identifies actionable targets for informing treatment decisions in metastatic breast cancer.

BACKGROUND: We report copy-number profiling by low-pass WGS (LP-WGS) in individual circulating tumour cells (CTCs) for guiding treatment in patients with metastatic breast cancer (MBC), comparing CTC results with mutations detected in circulating tumour DNA (ctDNA) in the same blood samples. METHODS: Across 10 patients with MBC who were progressing at the time of blood sampling and that had >20 CTCs detected by CellSearch®, 63 single cells (50 CTCs and 13 WBCs) and 16 cell pools (8 CTC pools and 8 WBC pools) were recovered from peripheral blood by CellSearch®/DEPArray™ and sequenced with Ampli1 LowPass technology (Menarini Silicon Biosystems). Copy-number aberrations were identified using the MSBiosuite software platform, and results were compared with mutations detected in matched plasma cfDNA analysed by targeted next-generation sequencing using the Oncomine™ Breast cfDNA Assay (Thermo Fisher). RESULTS: LP-WGS data demonstrated copy-number gains/losses in individual CTCs in regions including FGFR1, JAK2 and CDK6 in five patients, ERBB2 amplification in two HER2-negative patients and BRCA loss in two patients. Seven of eight matched plasmas also had mutations in ctDNA in PIK3CA, TP53, ESR1 and KRAS genes with mutant allele frequencies (MAF) ranging from 0.05 to 33.11%. Combining results from paired CTCs and ctDNA, clinically actionable targets were identified in all ten patients. CONCLUSION: This combined analysis of CTCs and ctDNA may offer a new approach for monitoring of disease progression and to direct therapy in patients with advanced MBC, at a time when they are coming towards the end of other treatment options.

Division and Transmission: Malaria Parasite Development in the Mosquito.

The malaria parasite life cycle alternates between two hosts: a vertebrate and the female Anopheles mosquito vector. Cell division, proliferation, and invasion are essential for parasite development, transmission, and survival. Most research has focused on Plasmodium development in the vertebrate, which causes disease; however, knowledge of malaria parasite development in the mosquito (the sexual and transmission stages) is now rapidly accumulating, gathered largely through investigation of the rodent malaria model, with Plasmodium berghei. In this review, we discuss the seminal genome-wide screens that have uncovered key regulators of cell proliferation, invasion, and transmission during Plasmodium sexual development. Our focus is on the roles of transcription factors, reversible protein phosphorylation, and molecular motors. We also emphasize the still-unanswered important questions around key pathways in cell division during the vector transmission stages and how they may be targeted in future studies.

PeMNet for Pectoral Muscle Segmentation.

As an important imaging modality, mammography is considered to be the global gold standard for early detection of breast cancer. Computer-Aided (CAD) systems have played a crucial role in facilitating quicker diagnostic procedures, which otherwise could take weeks if only radiologists were involved. In some of these CAD systems, breast pectoral segmentation is required for breast region partition from breast pectoral muscle for specific analysis tasks. Therefore, accurate and efficient breast pectoral muscle segmentation frameworks are in high demand. Here, we proposed a novel deep learning framework, which we code-named PeMNet, for breast pectoral muscle segmentation in mammography images. In the proposed PeMNet, we integrated a novel attention module called the Global Channel Attention Module (GCAM), which can effectively improve the segmentation performance of Deeplabv3+ using minimal parameter overheads. In GCAM, channel attention maps (CAMs) are first extracted by concatenating feature maps after paralleled global average pooling and global maximum pooling operation. CAMs are then refined and scaled up by multi-layer perceptron (MLP) for elementwise multiplication with CAMs in next feature level. By iteratively repeating this procedure, the global CAMs (GCAMs) are then formed and multiplied elementwise with final feature maps to lead to final segmentation. By doing so, CAMs in early stages of a deep convolution network can be effectively passed on to later stages of the network and therefore leads to better information usage. The experiments on a merged dataset derived from two datasets, INbreast and OPTIMAM, showed that PeMNet greatly outperformed state-of-the-art methods by achieving an IoU of 97.46%, global pixel accuracy of 99.48%, Dice similarity coefficient of 96.30%, and Jaccard of 93.33%, respectively.

Circulating Tumor DNA Profiling From Breast Cancer Screening Through to Metastatic Disease.

PURPOSE: We investigated the utility of the Oncomine Breast cfDNA Assay for detecting circulating tumor DNA (ctDNA) in women from a breast screening population, including healthy women with no abnormality detected by mammogram, and women on follow-up through to advanced breast cancer. MATERIALS AND METHODS: Blood samples were taken from 373 women (127 healthy controls recruited through breast screening, 28 ductal carcinoma in situ, 60 primary breast cancers, 47 primary breast cancer on follow-up, and 111 metastatic breast cancers [MBC]) to recover plasma and germline DNA for analysis with the Oncomine Breast cfDNA Assay on the Ion S5 platform. RESULTS: One hundred sixteen of 373 plasma samples had one or more somatic variants detected across eight of the 10 genes and were called ctDNA-positive; MBC had the highest proportion of ctDNA-positive samples (61; 55%) and healthy controls the lowest (20; 15.7%). ESR1, TP53, and PIK3CA mutations account for 93% of all variants detected and predict poor overall survival in MBC (hazard ratio = 3.461; 95% CI, 1.866 to 6.42; P = .001). Patients with MBC had higher plasma cell-free DNA levels, higher variant allele frequencies, and more polyclonal variants, notably in ESR1 than in all other groups. Only 15 individuals had evidence of potential clonal hematopoiesis of indeterminate potential mutations. CONCLUSION: We were able detect ctDNA across the breast cancer spectrum, notably in MBC where variants in ESR1, TP53, and PIK3CA predicted poor overall survival. The assay could be used to monitor emergence of resistance mutations such as in ESR1 that herald resistance to aromatase inhibitors to tailor adjuvant therapies. However, we suggest caution is needed when interpreting results from a single plasma sample as variants were also detected in a small proportion of HCs.

Improving ductal carcinoma in situ classification by convolutional neural network with exponential linear unit and rank-based weighted pooling.

Ductal carcinoma in situ (DCIS) is a pre-cancerous lesion in the ducts of the breast, and early diagnosis is crucial for optimal therapeutic intervention. Thermography imaging is a non-invasive imaging tool that can be utilized for detection of DCIS and although it has high accuracy (~ 88%), it is sensitivity can still be improved. Hence, we aimed to develop an automated artificial intelligence-based system for improved detection of DCIS in thermographs. This study proposed a novel artificial intelligence based system based on convolutional neural network (CNN) termed CNN-BDER on a multisource dataset containing 240 DCIS images and 240 healthy breast images. Based on CNN, batch normalization, dropout, exponential linear unit and rank-based weighted pooling were integrated, along with L-way data augmentation. Ten runs of tenfold cross validation were chosen to report the unbiased performances. Our proposed method achieved a sensitivity of 94.08 ± 1.22%, a specificity of 93.58 ± 1.49 and an accuracy of 93.83 ± 0.96. The proposed method gives superior performance than eight state-of-the-art approaches and manual diagnosis. The trained model could serve as a visual question answering system and improve diagnostic accuracy.

Protein phosphatase 1 regulates atypical mitotic and meiotic division in Plasmodium sexual stages.

PP1 is a conserved eukaryotic serine/threonine phosphatase that regulates many aspects of mitosis and meiosis, often working in concert with other phosphatases, such as CDC14 and CDC25. The proliferative stages of the malaria parasite life cycle include sexual development within the mosquito vector, with male gamete formation characterized by an atypical rapid mitosis, consisting of three rounds of DNA synthesis, successive spindle formation with clustered kinetochores, and a meiotic stage during zygote to ookinete development following fertilization. It is unclear how PP1 is involved in these unusual processes. Using real-time live-cell and ultrastructural imaging, conditional gene knockdown, RNA-seq and proteomic approaches, we show that Plasmodium PP1 is implicated in both mitotic exit and, potentially, establishing cell polarity during zygote development in the mosquito midgut, suggesting that small molecule inhibitors of PP1 should be explored for blocking parasite transmission.

Comparison of two targeted ultra-deep sequencing technologies for analysis of plasma circulating tumour DNA in endocrine-therapy-resistant breast cancer patients.

PURPOSE: There is growing interest in the application of circulating tumour DNA (ctDNA) as a sensitive tool for monitoring tumour evolution and guiding targeted therapy in patients with cancer. However, robust comparisons of different platform technologies are still required. Here we compared the InVisionSeq™ ctDNA Assay with the Oncomine™ Breast cfDNA Assay to assess their concordance and feasibility for the detection of mutations in plasma at low (< 0.5%) variant allele fraction (VAF). METHODS: Ninety-six plasma samples from 50 patients with estrogen receptor (ER)-positive metastatic breast cancer (mBC) were profiled using the InVision Assay. Results were compared to the Oncomine assay in 30 samples from 26 patients, where there was sufficient material and variants were covered by both assays. Longitudinal samples were analysed for 8 patients with endocrine resistance. RESULTS: We detected alterations in 59/96 samples from 34/50 patients analysed with the InVision assay, most frequently affecting ESR1, PIK3CA and TP53. Complete or partial concordance was found in 28/30 samples analysed by both assays, and VAF values were highly correlated. Excellent concordance was found for most genes, and most discordant calls occurred at VAF < 1%. In longitudinal samples from progressing patients with endocrine resistance, we detected consistent alterations in sequential samples, most commonly in ESR1 and PIK3CA. CONCLUSION: This study shows that both ultra-deep next-generation sequencing (NGS) technologies can detect genomic alternations even at low VAFs in plasma samples of mBC patients. The strong agreement of the technologies indicates sufficient reproducibility for clinical use as prognosic and predictive biomarker.

Modeling of diagnosis for metabolic syndrome by integrating symptoms into physiochemical indexes.

BACKGROUND: Metabolic syndrome (MS) is a major global health concern comprising a cluster of co-occurring conditions that increase the risk of heart disease, stroke and type 2 diabetes. MS is usually diagnosed using a combination of physiochemical indexes (such as BMI, abdominal circumference and blood pressure) but largely ignores clinical symptoms when investigating prevention and treatment of the disease. Exploring predictors of MS using multiple diagnostic indicators may improve early diagnosis and treatment of MS. Traditional Chinese medicine (TCM) attaches importance to the etiology of disease symptoms and indications using four diagnostic methods, which have long been used to treat metabolic disease. Therefore, in this study, we aimed to develop predictive indicators for MS using both physiochemical indexes and TCM methods. METHODS: Clinical information (including both physiochemical and TCM indexes) was obtained from a cohort of 586 individuals across 4 hospitals in China, comprising 136 healthy controls and 450 MS cases. Using this cohort, we compared three classic machine learning methods: decision tree (DT), support vector machine (SVM) and random forest (RF) towards MS diagnosis using physiochemical and TCM indexes, with the best model selected by comparing the accuracy, specificity and sensitivity of the three models. In parallel, the best proportional partition of the training data to the test data was confirmed by observing the changes in evaluation indexes using each model. Next, three subsets containing different categories of variables (including both TCM and physicochemical indexes combined - termed the "fused indexes", only physicochemical indexes, and TCM indexes only) were compared and analyzed using the best performing model and optimum training to test data proportion. Next, the best subset was selected through comprehensive comparative analysis, and then the important prediction variables were selected according to their weight. RESULTS: When comparing the three models, we found that the RF model had the highest average accuracy (average 0.942, 95%CI [0.925, 0.958]) and sensitivity (average 0.993, 95%CI [0.990, 0.996]). Besides, when the training set accounted for 80% of the cohort data, the specificity got the best value and the accuracy and sensitivity were also very high in RF model. In view of the performance of the three different subsets, the prediction accuracy and sensitivity of models analyzing the fused indexes and only physicochemical indexes remained at a high level. Further, the mean value of specificity of the model using fused indexes was 0.916, which was significantly higher than the model with only physicochemical indexes (average 0.822) and the model with only TCM indexes (average 0.403). Based on the RF model and data allocation ratio (8:2), we further extracted the top 20 most significant variables from the fused indexes, which included 14 physicochemical indexes and 6 TCM indexes including wiry pulse, chest tightness, spontaneous perspiration, greasy tongue coating etc. CONCLUSION: Compared with SVM and DT models, the RF model showed the best performance, especially when the ratio of the training set to test set is 8:2. Compared with single predictive indexes, the model constructed by combining physiochemical indexes with TCM indexes (i.e. the fused indexes) exhibited better predictive ability. In addition to common physicochemical indexes, some TCM indexes, such as wiry pulse, chest tightness, spontaneous perspiration, greasy tongue coating, can also improve diagnosis of MS.

COVID-19 classification by CCSHNet with deep fusion using transfer learning and discriminant correlation analysis.

AIM: : COVID-19 is a disease caused by a new strain of coronavirus. Up to 18th October 2020, worldwide there have been 39.6 million confirmed cases resulting in more than 1.1 million deaths. To improve diagnosis, we aimed to design and develop a novel advanced AI system for COVID-19 classification based on chest CT (CCT) images. METHODS: : Our dataset from local hospitals consisted of 284 COVID-19 images, 281 community-acquired pneumonia images, 293 secondary pulmonary tuberculosis images; and 306 healthy control images. We first used pretrained models (PTMs) to learn features, and proposed a novel (L, 2) transfer feature learning algorithm to extract features, with a hyperparameter of number of layers to be removed (NLR, symbolized as L). Second, we proposed a selection algorithm of pretrained network for fusion to determine the best two models characterized by PTM and NLR. Third, deep CCT fusion by discriminant correlation analysis was proposed to help fuse the two features from the two models. Micro-averaged (MA) F1 score was used as the measuring indicator. The final determined model was named CCSHNet. RESULTS: : On the test set, CCSHNet achieved sensitivities of four classes of 95.61%, 96.25%, 98.30%, and 97.86%, respectively. The precision values of four classes were 97.32%, 96.42%, 96.99%, and 97.38%, respectively. The F1 scores of four classes were 96.46%, 96.33%, 97.64%, and 97.62%, respectively. The MA F1 score was 97.04%. In addition, CCSHNet outperformed 12 state-of-the-art COVID-19 detection methods. CONCLUSIONS: : CCSHNet is effective in detecting COVID-19 and other lung infectious diseases using first-line clinical imaging and can therefore assist radiologists in making accurate diagnoses based on CCTs.

Patient acceptability of circulating tumour DNA testing in endometrial cancer follow-up.

OBJECTIVE: Circulating tumour DNA (ctDNA) is emerging as a potential option to detect disease recurrence in many cancer types, however, ensuring patient acceptability of changing clinical practice and the introduction of new technology is paramount. METHODS: Patients enrolled in a non-intervention cohort study determining the ability of ctDNA to detect recurrent endometrial cancer (EC) were invited to participate in a semi-structured interview. Analysis was performed by Template Analysis. RESULTS: Eighteen patients were interviewed. A ctDNA blood test was viewed by participants as more physically and psychologically acceptable than clinical examination to monitor for EC recurrence. In particular, participants expressed overwhelming preference for a blood test rather than pelvic examination. Although participants acknowledged that an abnormal ctDNA result could cause anxiety, they expressed a preference to be informed of their results, even if a recurrence was too small to detect radiologically. Explanations for these opinions were a desire for certainty whether their cancer would recur or not, and knowledge would help them be more aware of symptoms that should be reported to their clinician. CONCLUSIONS: ctDNA monitoring to identify EC recurrence appears to be acceptable to patients, and for many, it may be preferable to clinical examination.

A Heuristic Neural Network Structure Relying on Fuzzy Logic for Images Scoring.

Traditional deep learning methods are sub-optimal in classifying ambiguity features, which often arise in noisy and hard to predict categories, especially, to distinguish semantic scoring. Semantic scoring, depending on semantic logic to implement evaluation, inevitably contains fuzzy description and misses some concepts, for example, the ambiguous relationship between normal and probably normal always presents unclear boundaries (normal - more likely normal - probably normal). Thus, human error is common when annotating images. Differing from existing methods that focus on modifying kernel structure of neural networks, this study proposes a dominant fuzzy fully connected layer (FFCL) for Breast Imaging Reporting and Data System (BI-RADS) scoring and validates the universality of this proposed structure. This proposed model aims to develop complementary properties of scoring for semantic paradigms, while constructing fuzzy rules based on analyzing human thought patterns, and to particularly reduce the influence of semantic conglutination. Specifically, this semantic-sensitive defuzzier layer projects features occupied by relative categories into semantic space, and a fuzzy decoder modifies probabilities of the last output layer referring to the global trend. Moreover, the ambiguous semantic space between two relative categories shrinks during the learning phases, as the positive and negative growth trends of one category appearing among its relatives were considered. We first used the Euclidean Distance (ED) to zoom in the distance between the real scores and the predicted scores, and then employed two sample t test method to evidence the advantage of the FFCL architecture. Extensive experimental results performed on the CBIS-DDSM dataset show that our FFCL structure can achieve superior performances for both triple and multiclass classification in BI-RADS scoring, outperforming the state-of-the-art methods.

ResNet-SCDA-50 for Breast Abnormality Classification.

(Aim) Breast cancer is the most common cancer in women and the second most common cancer worldwide. With the rapid advancement of deep learning, the early stages of breast cancer development can be accurately detected by radiologists with the help of artificial intelligence systems. (Method) Based on mammographic imaging, a mainstream clinical breast screening technique, we present a diagnostic system for accurate classification of breast abnormalities based on ResNet-50. To improve the proposed model, we created a new data augmentation framework called SCDA (Scaling and Contrast limited adaptive histogram equalization Data Augmentation). In its procedure, we first conduct the scaling operation to the original training set, followed by applying contrast limited adaptive histogram equalisation (CLAHE) to the scaled training set. By stacking the training set after SCDA with the original training set, we formed a new training set. The network trained by the augmented training set, was coined as ResNet-SCDA-50. Our system, which aims at a binary classification on mammographic images acquired from INbreast and MINI-MIAS, classifies masses, microcalcification as "abnormal", while normal regions are classified as "normal". (Results) We present the first attempt to use the image contrast enhancement method as the data augmentation method, resulting in an averaged 98.55 percent specificity and 92.83 percent sensitivity, which gives our best model an overall accuracy of 95.74 percent. (Conclusion) Our proposed method is effective in classifying breast abnormality.