Diagnostics

Smartphone-based devices and applications (SBDAs) with cost effectiveness and remote sensing are the most promising and effective means of delivering mobile healthcare (mHealthcare). Several SBDAs have been commercialized for the personalized monitoring and/or management of basic physiological parameters, such as blood pressure, weight, body analysis, pulse rate, electrocardiograph, blood glucose, blood glucose saturation, sleeping and physical activity. With advances in Bluetooth technology, software, cloud computing and remote sensing, SBDAs provide real-time on-site analysis and telemedicine opportunities in remote areas. This scenario is of utmost importance for developing countries, where the number of smartphone users is about 70% of 6.8 billion cell phone subscribers worldwide with limited access to basic healthcare service. The technology platform facilitates patient-doctor communication and the patients to effectively manage and keep track of their medical conditions. Besides tremendous healthcare cost savings, SBDAs are very critical for the monitoring and effective management of emerging epidemics and food contamination outbreaks. The next decade will witness pioneering advances and increasing applications of SBDAs in this exponentially growing field of mHealthcare. This article provides a critical review of commercial SBDAs that are being widely used for personalized healthcare monitoring and management.

Over the past decade, convolutional neural networks (CNN) have shown very competitive performance in medical image analysis tasks, such as disease classification, tumor segmentation, and lesion detection. CNN has great advantages in extracting local features of images. However, due to the locality of convolution operation, it cannot deal with long-range relationships well. Recently, transformers have been applied to computer vision and achieved remarkable success in large-scale datasets. Compared with natural images, multi-modal medical images have explicit and important long-range dependencies, and effective multi-modal fusion strategies can greatly improve the performance of deep models. This prompts us to study transformer-based structures and apply them to multi-modal medical images. Existing transformer-based network architectures require large-scale datasets to achieve better performance. However, medical imaging datasets are relatively small, which makes it difficult to apply pure transformers to medical image analysis. Therefore, we propose TransMed for multi-modal medical image classification. TransMed combines the advantages of CNN and transformer to efficiently extract low-level features of images and establish long-range dependencies between modalities. We evaluated our model on two datasets, parotid gland tumors classification and knee injury classification. Combining our contributions, we achieve an improvement of 10.1% and 1.9% in average accuracy, respectively, outperforming other state-of-the-art CNN-based models. The results of the proposed method are promising and have tremendous potential to be applied to a large number of medical image analysis tasks. To our best knowledge, this is the first work to apply transformers to multi-modal medical image classification.

Manual diagnosis of skin cancer is time-consuming and expensive; therefore, it is essential to develop automated diagnostics methods with the ability to classify multiclass skin lesions with greater accuracy. We propose a fully automated approach for multiclass skin lesion segmentation and classification by using the most discriminant deep features. First, the input images are initially enhanced using local color-controlled histogram intensity values (LCcHIV). Next, saliency is estimated using a novel Deep Saliency Segmentation method, which uses a custom convolutional neural network (CNN) of ten layers. The generated heat map is converted into a binary image using a thresholding function. Next, the segmented color lesion images are used for feature extraction by a deep pre-trained CNN model. To avoid the curse of dimensionality, we implement an improved moth flame optimization (IMFO) algorithm to select the most discriminant features. The resultant features are fused using a multiset maximum correlation analysis (MMCA) and classified using the Kernel Extreme Learning Machine (KELM) classifier. The segmentation performance of the proposed methodology is analyzed on ISBI 2016, ISBI 2017, ISIC 2018, and PH2 datasets, achieving an accuracy of 95.38%, 95.79%, 92.69%, and 98.70%, respectively. The classification performance is evaluated on the HAM10000 dataset and achieved an accuracy of 90.67%. To prove the effectiveness of the proposed methods, we present a comparison with the state-of-the-art techniques.

1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) alone, and in combination with N-hydroxysuccinimide (NHS) or sulfoNHS were employed for crosslinking anti-human fetuin A (HFA) antibodies on 3-aminopropyltriethoxysilane (APTES)-functionalized surface plasmon resonance (SPR) gold chip and 96-well microtiter plate. The SPR immunoassay and sandwich enzyme linked immunosorbent immunoassay (ELISA) for HFA clearly demonstrated that EDC crosslinks anti-HFA antibodies to APTES-functionalized bioanalytical platforms more efficiently than EDC/NHS and EDC/sulfoNHS at a normal pH of 7.4. Similar results were obtained by sandwich ELISAs for human Lipocalin-2 and human albumin, and direct ELISA for horseradish peroxidase. The more efficient crosslinking of antibodies by EDC to the APTES-functionalized platforms increased the cost-effectiveness and analytical performance of our immunoassays. This study will be of wide interest to researchers developing immunoassays on APTES-functionalized platforms that are being widely used in biomedical diagnostics, biosensors, lab-on-a-chip and point-of-care-devices. It stresses a critical need of an intensive investigation into the mechanisms of EDC-based amine-carboxyl coupling under various experimental conditions.

In only a few months after initial discovery in Wuhan, China, SARS-CoV-2 and the associated coronavirus disease 2019 (COVID-19) have become a global pandemic causing significant mortality and morbidity and implementation of strict isolation measures. In the absence of vaccines and effective therapeutics, reliable serological testing must be a key element of public health policy to control further spread of the disease and gradually remove quarantine measures. Serological diagnostic tests are being increasingly used to provide a broader understanding of COVID-19 incidence and to assess immunity status in the population. However, there are discrepancies between claimed and actual performance data for serological diagnostic tests on the market. In this study, we conducted a review of independent studies evaluating the performance of SARS-CoV-2 serological tests. We found significant variability in the accuracy of marketed tests and highlight several lab-based and point-of-care rapid serological tests with high levels of performance. The findings of this review highlight the need for ongoing independent evaluations of commercialized COVID-19 diagnostic tests.

Ovarian cancer is the most-deadly gynecologic malignancy, with greater than 14,000 women expected to succumb to the disease this year in the United States alone. In the front-line setting, patients are treated with a platinum and taxane doublet. Although 40–60% of patients achieve complete clinical response to first-line chemotherapy, 25% are inherently platinum-resistant or refractory with a median overall survival of about one year. More than 80% of women afflicted with ovarian cancer will recur. Many attempts have been made to understand the mechanism of platinum and taxane based chemotherapy resistance. However, despite decades of research, few predictive markers of chemotherapy resistance have been identified. Here, we review the current understanding of one of the most common genetic alterations in epithelial ovarian cancer, CCNE1 (cyclin E1) amplification, and its role as a potential predictive marker of cytotoxic chemotherapy resistance. CCNE1 amplification has been identified as a primary oncogenic driver in a subset of high grade serous ovarian cancer that have an unmet clinical need. Understanding the interplay between cyclin E1 amplification and other common ovarian cancer genetic alterations provides the basis for chemotherapeutic resistance in CCNE1 amplified disease. Exploration of the effect of cyclin E1 amplification on the cellular machinery that causes dysregulated proliferation in cancer cells has allowed investigators to explore promising targeted therapies that provide the basis for emerging clinical trials.

Breast cancer is becoming more dangerous by the day. The death rate in developing countries is rapidly increasing. As a result, early detection of breast cancer is critical, leading to a lower death rate. Several researchers have worked on breast cancer segmentation and classification using various imaging modalities. The ultrasonic imaging modality is one of the most cost-effective imaging techniques, with a higher sensitivity for diagnosis. The proposed study segments ultrasonic breast lesion images using a Dilated Semantic Segmentation Network (Di-CNN) combined with a morphological erosion operation. For feature extraction, we used the deep neural network DenseNet201 with transfer learning. We propose a 24-layer CNN that uses transfer learning-based feature extraction to further validate and ensure the enriched features with target intensity. To classify the nodules, the feature vectors obtained from DenseNet201 and the 24-layer CNN were fused using parallel fusion. The proposed methods were evaluated using a 10-fold cross-validation on various vector combinations. The accuracy of CNN-activated feature vectors and DenseNet201-activated feature vectors combined with the Support Vector Machine (SVM) classifier was 90.11 percent and 98.45 percent, respectively. With 98.9 percent accuracy, the fused version of the feature vector with SVM outperformed other algorithms. When compared to recent algorithms, the proposed algorithm achieves a better breast cancer diagnosis rate.

Breast density estimation with visual evaluation is still challenging due to low contrast and significant fluctuations in the mammograms’ fatty tissue background. The primary key to breast density classification is to detect the dense tissues in the mammographic images correctly. Many methods have been proposed for breast density estimation; nevertheless, most of them are not fully automated. Besides, they have been badly affected by low signal-to-noise ratio and variability of density in appearance and texture. This study intends to develop a fully automated and digitalized breast tissue segmentation and classification using advanced deep learning techniques. The conditional Generative Adversarial Networks (cGAN) network is applied to segment the dense tissues in mammograms. To have a complete system for breast density classification, we propose a Convolutional Neural Network (CNN) to classify mammograms based on the standardization of Breast Imaging-Reporting and Data System (BI-RADS). The classification network is fed by the segmented masks of dense tissues generated by the cGAN network. For screening mammography, 410 images of 115 patients from the INbreast dataset were used. The proposed framework can segment the dense regions with an accuracy, Dice coefficient, Jaccard index of 98%, 88%, and 78%, respectively. Furthermore, we obtained precision, sensitivity, and specificity of 97.85%, 97.85%, and 99.28%, respectively, for breast density classification. This study’s findings are promising and show that the proposed deep learning-based techniques can produce a clinically useful computer-aided tool for breast density analysis by digital mammography.

An electrochemical immunosensor modified with the streptavidin/biotin system on screen printed carbon electrodes (SPCEs) for the detection of the dengue NS1 antigen was developed in this study. Monoclonal anti-NS1 capture antibody was immobilized on streptavidin-modified SPCEs to increase the sensitivity of the assay. Subsequently, a direct sandwich enzyme linked immunosorbent assay (ELISA) format was developed and optimized. An anti-NS1 detection antibody conjugated with horseradish peroxidase enzyme (HRP) and 3,3,5,5'-tetramethybezidine dihydrochloride (TMB/H2O2) was used as an enzyme mediator. Electrochemical detection was conducted using the chronoamperometric technique, and electrochemical responses were generated at −200 mV reduction potential. The calibration curve of the immunosensor showed a linear response between 0.5 µg/mL and 2 µg/mL and a detection limit of 0.03 µg/mL. Incorporation of a streptavidin/biotin system resulted in a well-oriented antibody immobilization of the capture antibody and consequently enhanced the sensitivity of the assay. In conclusion, this immunosensor is a promising technology for the rapid and convenient detection of acute dengue infection in real serum samples.

Pseudomonas aeruginosa (PA) is a common bacterial pathogen in chronic wounds known for its propensity to form biofilms and evade conventional treatment methods. Early detection of PA in wounds is critical to the mitigation of more severe wound outcomes. Point-of-care bacterial fluorescence imaging illuminates wounds with safe, violet light, triggering the production of cyan fluorescence from PA. A prospective single blind clinical study was conducted to determine the positive predictive value (PPV) of cyan fluorescence for the detection of PA in wounds. Bacterial fluorescence using the MolecuLight i:X imaging device revealed cyan fluorescence signal in 28 chronic wounds, including venous leg ulcers, surgical wounds, diabetic foot ulcers and other wound types. To correlate the cyan signal to the presence of PA, wound regions positive for cyan fluorescence were sampled via curettage. A semi-quantitative culture analysis of curettage samples confirmed the presence of PA in 26/28 wounds, resulting in a PPV of 92.9%. The bacterial load of PA from cyan-positive regions ranged from light to heavy. Less than 20% of wounds that were positive for PA exhibited the classic symptoms of PA infection. These findings suggest that cyan detected on fluorescence images can be used to reliably predict bacteria, specifically PA at the point-of-care.