Search

Purpose: To train deep learning models to differentiate benign and malignant breast tumors in ultrasound images, we need to collect many training samples with clear labels. In general, biopsy results can be used as benign/malignant labels. However, most clinical samples generally do not have biopsy results. Previous works have proposed generating benign/malignant labels according to Breast Imaging, Reporting and Data System (BI‐RADS) ratings. However, this approach will cause noisy labels, which means that the benign/malignant labels produced from BI‐RADS diagnoses may be inconsistent with the ground truths. Consequently, deep models will overfit the noisy labels and hence obtain poor generalization performance. In this work, we mainly focus on how to reduce the negative effect of noisy labels when they are used to train breast tumor classification models. Methods: We propose an effective approach called noise filter network (NF‐Net) to address the problem of noisy labels when training breast tumor classification models. Specifically, to prevent deep models from overfitting the noisy labels, we propose incorporating two softmax layers for classification. Additionally, to strengthen the effect of clean labels, we design a teacher–student module for distilling the knowledge of clean labels. Results: We conduct extensive comparisons with the existing works on addressing noisy labels. Our method achieves a classification accuracy of 73%, with a precision of 69%, recall of 80%, and F1‐score of 0.74. This result is significantly better than those of the existing state‐of‐the‐art works on addressing noisy labels. Conclusions: This work provides a means to overcome the label shortage problem in training breast tumor classification models. Specifically, we can generate benign/malignant labels according to the BI‐RADS ratings. Although this approach will cause noisy labels, the design of NF‐Net can effectively reduce the negative effect of such labels. [ABSTRACT FROM AUTHOR]

Foodborne illnesses associated with foodborne pathogens Listeria monocytogenes and Staphylococcus aureus occur at high frequencies. In this study, species‐specific primers and probes sets for the detection of L. monocytogenes and S. aureus were designed based on the nucleotide sequences of the actin‐assembly inducing protein precursor (actA) and clumping factor A (clfA) genes, respectively. We established a duplex RPA assay that is capable of simultaneously detecting L. monocytogenes and S. aureus within 15 min. With the plasmid standard, the sensitivity of the double RPA to L. monocytogenes and S. aureus can reach 102 copies/μL. The method also detected DNA extracted from samples inoculated with L. monocytogenes and S. aureus. After 4 hr of incubation, the sensitivity of the food model to L. monocytogenes and S. aureus reached 230 and 270 cfu/reaction, respectively. The assay was highly specific and had no cross‐reaction between closely related bacteria. Further assessment of this method on strains isolated within the lab collection showed that the duplex RPA provided the same testing results as microbiological methods. Therefore, the assay is effective for the rapid detection and preliminary screening of L. monocytogenes and S. aureus. Practical applications: Listeria monocytogenes and Staphylococcus aureus are two major foodborne pathogens and have been found in a wide range of foods. We developed a duplex RPA specific targeting actA and clfA genes with high‐amplification efficiency. We evaluated the performance of duplex RPA by building a food model. The detection results by the duplex RPA were compared with PCR and no significant differences were found. This study provides the information of the developed duplex RPA for detecting two specific foodborne pathogens, which will be helpful for further studies about simultaneous detection of several targets using duplex RPA. [ABSTRACT FROM AUTHOR]