(1) Purpose: To improve the capability of EfficientNet, including developing a cropping method called Random Center Cropping (RCC) to retain the original image resolution and significant features on the images’ center area, reducing the downsampling scale of EfficientNet to facilitate the small resolution images of RPCam datasets, and integrating attention and Feature Fusion (FF) mechanisms with EfficientNet to obtain features containing rich semantic information. (2) Methods: We adopt the Convolutional Neural Network (CNN) to detect and classify lymph node metastasis in breast cancer. (3) Results: Experiments illustrate that our methods significantly boost performance of basic CNN architectures, where the best-performed method achieves an accuracy of 97.96% ± 0.03% and an Area Under the Curve (AUC) of 99.68% ± 0.01% on RPCam datasets, respectively. (4) Conclusions: (1) To our limited knowledge, we are the only study to explore the power of EfficientNet on Metastatic Breast Cancer (MBC) classification, and elaborate experiments are conducted to compare the performance of EfficientNet with other state-of-the-art CNN models. It might provide inspiration for researchers who are interested in image-based diagnosis using Deep Learning (DL). (2) We design a novel data augmentation method named RCC to promote the data enrichment of small resolution datasets. (3) All of our four technological improvements boost the performance of the original EfficientNet.

Lipid-based nanoparticles (LBNPs) are biocompatible and biodegradable vesicles that are considered to be one of the most efficient drug delivery platforms. Due to the prominent advantages, such as long circulation time, slow drug release, reduced toxicity, high transfection efficiency, and endosomal escape capacity, such synthetic nanoparticles have been widely used for carrying genetic therapeutics, particularly nucleic acids that can be applied in the treatment for various diseases, including congenital diseases, cancers, virus infections, and chronic inflammations. Despite great merits and multiple successful applications, many extracellular and intracellular barriers remain and greatly impair delivery efficacy and therapeutic outcomes. As such, the current state of knowledge and pitfalls regarding the gene delivery and construction of LBNPs will be initially summarized. In order to develop a new generation of LBNPs for improved delivery profiles and therapeutic effects, the modification strategies of LBNPs will be reviewed. On the basis of these developed modifications, the performance of LBNPs as therapeutic nanoplatforms have been greatly improved and extensively applied in immunotherapies, including infectious diseases and cancers. However, the therapeutic applications of LBNPs systems are still limited due to the undesirable endosomal escape, potential aggregation, and the inefficient encapsulation of therapeutics. Herein, we will review and discuss recent advances and remaining challenges in the development of LBNPs for nucleic acid-based immunotherapy.

Background: Parkinson’s disease (PD) is one of the most common neurological disorders that can severely affect the ability to perform daily activities. The clinical presentation of PD includes motor and nonmotor symptoms. The motor symptoms generally involve movement conditions like tremors, rigidity, slowness, and impaired balance. In contrast, the nonmotor symptoms are often not apparent but can affect various organ systems, such as the urinary and gastrointestinal systems, and mental health. Gene mutations and toxic environmental factors have contributed significantly to PD; nevertheless, its cause and underlying mechanism remain unknown. Currently, treatments such as dopamine agonists, RNA molecules, and antioxidants can, to some extent, alleviate the motor symptoms triggered by PD. However, these medicines cannot effectively halt ongoing dopaminergic damage, mainly because the blood-brain barrier (BBB) lowers the efficiency of drug delivery. Recently, extracellular vesicles (EVs), a novel drug delivery platform, have been widely used in various neurological diseases, including stroke and brain tumors, because of their excellent biocompatibility, their ability to penetrate the BBB without toxicity, and their target specificity. EVs thus provide a promising therapeutic for treating PD. Objective: This review focuses on novel therapies based on EVs in practice. Herein, we briefly introduce the biogenesis, composition, isolation, and characterization of EVs, and we discuss strategies for loading therapeutic agents onto EVs and recent applications for PD treatment. Moreover, we discuss perspectives on the direction of preclinical and clinical studies regarding novel and effective therapies. Methods: A literature search regarding PD treatment based on extracellular vesicles was performed in PubMed (updated in June 2020). Treatment, therapy, drug delivery, extracellular vesicles, and their combinations were the search queries. Both systematic reviews and original publications were included. Searched results were selected and compared based on relevance. Articles published in the last five years were given top priority. Conclusion: PD is a heterogeneous disease that can be treated by using pharmacologic approaches (e.g., dopamine agonists and levodopa) and nonpharmacologic approaches (e.g., music), based on symptoms and progression level in patients. Even though current treatments have demonstrated effectiveness, clinical challenges remain because the BBB reduces medication received and lowers the efficacy of drug delivery, which impairs the treatment’s effect. Therefore, EVs, as an emerging delivery platform, are highly promising for PD treatment since they can readily cross the BBB with high therapeutic efficiency through the loading or functionalization process. However, defining a safe source of EVs, reliably purifying and isolating EVs with high yield, and improving the efficacy of therapeutic loading in EVs remain challenging in this field. Therefore, future investigations should focus on generating large-scale exosomal carriers and designing new effective drugs encapsulated in EVs for better efficacy.

The identification of profiled cancer‐related genes plays an essential role in cancer diagnosis and treatment. Based on literature research, the classification of genetic mutations continues to be done manually nowadays. Manual classification of genetic mutations is pathologist‐dependent, subjective, and time‐consuming. To improve the accuracy of clinical interpretation, scientists have proposed computational‐based approaches for automatic analysis of mutations with the advent of next‐generation sequencing technologies. Nevertheless, some challenges, such as multiple classifications, the complexity of texts, redundant descriptions, and inconsistent interpretation, have limited the development of algorithms. To overcome these difficulties, we have adapted a deep learning method named Bidirectional Encoder Representations from Transformers (BERT) to classify genetic mutations based on text evidence from an annotated database. During the training, three challenging features such as the extreme length of texts, biased data presentation, and high repeatability were addressed. Finally, the BERT+abstract demonstrates satisfactory results with 0.80 logarithmic loss, 0.6837 recall, and 0.705 F‐measure. It is feasible for BERT to classify the genomic mutation text within literature‐based datasets. Consequently, BERT is a practical tool for facilitating and significantly speeding up cancer research towards tumor progression, diagnosis, and the design of more precise and effective treatments.

The Gleason grading system using histological images is the most powerful diagnostic and prognostic predictor of prostate cancer. The current standard inspection is evaluating Gleason H&E-stained histopathology images by pathologists. However, it is complicated, time-consuming, and subject to observers. Deep learning (DL) based-methods that automatically learn image features and achieve higher generalization ability have attracted significant attention. However, challenges remain especially using DL to train the whole slide image (WSI), a predominant clinical source in the current diagnostic setting, containing billions of pixels, morphological heterogeneity, and artifacts. Hence, we proposed a convolutional neural network (CNN)-based automatic classification method for accurate grading of PCa using whole slide histopathology images. In this paper, a data augmentation method named Patch-Based Image Reconstruction (PBIR) was proposed to reduce the high resolution and increase the diversity of WSIs. In addition, a distribution correction (DC) module was developed to enhance the adaption of pretrained model to the target dataset by adjusting the data distribution. Besides, a Quadratic Weighted Mean Square Error (QWMSE) function was presented to reduce the misdiagnosis caused by equal Euclidean distances. Our experiments indicated the combination of PBIR, DC, and QWMSE function was necessary for achieving superior expert-level performance, leading to the best results (0.8885 quadratic-weighted kappa coefficient).