Sensors

A Self-Calibrating IoT Portable Electrochemical Immunosensor for Serum Human Epididymis Protein 4 as a Tumor Biomarker for Ovarian Cancer

Nowadays, analytical techniques are moving towards the development of smart biosensing strategies for the point-of-care accurate screening of disease biomarkers, such as human epididymis protein 4 (HE4), a recently discovered serum marker for early ovarian cancer diagnosis. In this context, the present work represents the first implementation of a competitive enzyme-labelled magneto-immunoassay exploiting a homemade IoT Wi-Fi cloud-based portable potentiostat for differential pulse voltammetry readout. The electrochemical device was specifically designed to be capable of autonomous calibration and data processing, switching between calibration, and measurement modes: in particular, firstly, a baseline estimation algorithm is applied for correct peak computation, then calibration function is built by interpolating data with a four-parameter logistic function. The calibration function parameters are stored on the cloud for inverse prediction to determine the concentration of unknown samples. Interpolation function calibration and concentration evaluation are performed directly on-board, thus reducing the power consumption. The analytical device was validated in human serum, demonstrating good sensing performance for analysis of HE4 with detection and quantitation limits in human serum of 3.5 and 29.2 pM, respectively, reaching the sensitivity that is required for diagnostic purposes, with high potential for applications as portable and smart diagnostic tool for point-of-care testing.

Multimodal MRI Image Fusion for Early Automatic Staging of Endometrial Cancer

This magnetic resonance imaging multimodal fusion study aims to automate the staging of endometrial cancer using deep learning and to compare the diagnostic performance of deep learning with that of radiologists in the staging of endometrial cancer. This study retrospectively investigated 122 patients with pathologically confirmed early EC from January 1, 2025 to December 31, 2021. Of these patients, 68 were in the International Federation of Gynecology and Obstetrics (FIGO) stage IA, and 54 were in FIGO stage IB. Based on the Swin transformer model and its proprietary SW-MSA (shift window multiple self-coherence) module, magnetic resonance imaging (MRI) images in each of the three planes (sagittal, coronal, and transverse) are cropped, enhanced, and classified, and fusion experiments in the three planes are performed simultaneously. Selecting one plane for the experiment, the accuracy of IA and IB classification was 0.988 in the sagittal, 0.96 in the coronal, and 0.94 in the transverse position, and classification accuracy after the fusion of three planes reached 1. Finally, the automatic classification method based on the Swin transformer has an accuracy of 1, a recall of 1, and a specificity of 1 for early EC classification. In this study, the multimodal fusion approach accurately classified early EC. It was comparable to what a radiologist would perform and simpler and more precise than previous methods that required segmenting followed by staging.

Design and Characterization of a Hyperspectral Colposcope Based on Dual-LCTF VNIR Narrow-Band Illumination

Early detection of precancerous cervical lesions is critical for improving patient management and clinical outcomes. Hyperspectral imaging has emerged as a promising non-invasive, label-free imaging modality for rapid medical diagnosis. This work presents the development of a liquid-crystal-tunable-filter-based hyperspectral colposcopy system covering the visible and near-infrared spectral ranges. The proposed system integrates two tunable filters into an existing Optomic OP-C5 clinical colposcope, enabling hyperspectral acquisition from 460 to 1000 nm with 130 spectral bands at 5 nm resolution using a panchromatic camera. Two alternative acquisition strategies were investigated: (i) filtering the light received by the system, or (ii) filtering the light emitted toward the sample. In addition, wavelength-dependent exposure control was studied to compensate for reduced system sensitivity and improve the signal-to-noise ratio in low-efficiency spectral regions. The system was benchmarked against a previous custom hyperspectral implementation based on a commercial camera. The comparative analysis highlights the advantages and limitations of both approaches, demonstrating the proposed system’s suitability for integration into clinical workflows and its potential for early detection of precancerous cervical lesions during routine colposcopic examinations.

A Dual Immunosensor Based on Optical Weak Value Amplification for Simultaneous Detection of CA125 and HE4

Simultaneous detection of multiple biomarkers is essential for effective cancer screening. Taking ovarian cancer as an example, the combined detection of CA125 and HE4 has proven to be the most efficient and accurate among multiple biomarker combinations. In this study, we proposed a dual immunosensor based on weak value amplification (WVA) to detect ovarian cancer. By modifying the sensor surface through a self-assembled monolayer technique and utilizing recombinant protein G for antibody enrichment and directional capture, the sensor enables high-precision, simultaneous detection of CA125 and HE4, with detection limits of 5.39 U/mL and 1.79 ng/mL, respectively. Furthermore, the sensor demonstrates excellent specificity, effectively distinguishing target analytes from non-target molecules. This study provides a novel approach for early cancer screening and clinical diagnosis, highlighting the potential of WVA-based immunosensors in ovarian cancer detection.

Advanced Feature Extraction for Cervical Cancer Image Classification: Integrating Neural Feature Extraction and AutoInt Models

Cervical cancer remains a significant global public health challenge, affecting over half a million women annually, with a mortality rate of approximately 60%, especially in resource-limited regions. This study presents an advanced methodology for cervical cancer diagnosis through deep learning techniques. Utilizing a publicly available cervical cancer image dataset, the research introduces a novel classification framework that integrates a Neural Feature Extractor (NFE) based on a pre-trained VGG16 architecture and an AutoInt model for automatic feature interaction learning. The extracted features are processed through machine learning classifiers such as KNN, LGBM, Extra Trees, and others for classification tasks. Among these classifiers, KNN achieved the highest accuracy of 99.96%, followed closely by LGBM at 99.92%. This study also assesses the computational complexity of various models, demonstrating that simpler models like LDA exhibit faster prediction times, while more complex models, such as KNN and LGBM, provide higher accuracy. These findings highlight the potential of deep learning frameworks in improving cervical cancer classification accuracy, especially in resource-limited environments.

Role of Defects of Carbon Nanomaterials in the Detection of Ovarian Cancer Cells in Label-Free Electrochemical Immunosensors

Developing label-free immunosensors to detect ovarian cancer (OC) by cancer antigen (CA125) is essential to improving diagnosis and protecting women from life-threatening diseases. Four types of carbon nanomaterials, such as multi-wall carbon nanotubes (MWCNTs), vapor-grown carbon fiber (VGCFs), graphite KS4, and carbon black super P (SP), have been treated with acids to prepare a carbon nanomaterial/gold (Au) nanocomposite. The AuNPs@carbon nanocomposite was electrochemically deposited on a glassy carbon electrode (GCE) to serve as a substrate to fabricate a label-free immunosensor for the detection of CA125. Among the four AuNPs@carbon composite, the AuNPs@MWCNTs-based sensor exhibited a high sensitivity of 0.001 µg/mL for the biomarker CA125 through the square wave voltammetry (SWV) technique. The high conductivity and surface area of MWCNTs supported the immobilization of AuNPs. Moreover, the carboxylic (COO-) functional groups in MWCNT improved to a higher quantity after the acid treatment, which served as an excellent support for the fabrication of electrochemical biosensors. The present method aims to explore an environmentally friendly synthesis of a layer-by-layer (LBL) assembly of AuNPs@carbon nanomaterials electrochemical immunoassay to CA125 in a clinical diagnosis at a low cost and proved feasible for point-of-care diagnosis.

Full-Automatic High-Efficiency Mueller Matrix Microscopy Imaging for Tissue Microarray Inspection

This paper proposes a full-automatic high-efficiency Mueller matrix microscopic imaging (MMMI) system based on the tissue microarray (TMA) for cancer inspection for the first time. By performing a polar decomposition on the sample’s Mueller matrix (MM) obtained by a transmissive MMMI system we established, the linear phase retardance equivalent waveplate fast-axis azimuth and the linear phase retardance are obtained for distinguishing the cancerous tissues from the normal ones based on the differences in their polarization characteristics, where three analyses methods including statistical analysis, the gray-level co-occurrence matrix analysis (GLCM) and the Tamura image processing method (TIPM) are used. Previous MMMI medical diagnostics typically utilized discrete slices for inspection under a high-magnification objective (20×–50×) with a small field of view, while we use the TMA under a low-magnification objective (5×) with a large field of view. Experimental results indicate that MMMI based on TMA can effectively analyze the pathological variations in biological tissues, inspect cancerous cervical tissues, and thus contribute to the diagnosis of postoperative cancer biopsies. Such an inspection method, using a large number of samples within a TMA, is beneficial for obtaining consistent findings and good reproducibility.

Electrochemical and Photoelectrochemical Immunosensors for the Detection of Ovarian Cancer Biomarkers

Photoelectrochemical (PEC) sensing is an emerging technological innovation for monitoring small substances/molecules in biological or non–biological systems. In particular, there has been a surge of interest in developing PEC devices for determining molecules of clinical significance. This is especially the case for molecules that are markers for serious and deadly medical conditions. The increased interest in PEC sensors to monitor such biomarkers can be attributed to the many apparent advantages of the PEC system, including an enhanced measurable signal, high potential for miniaturization, rapid testing, and low cost, amongst others. The growing number of published research reports on the subject calls for a comprehensive review of the various findings. This article is a review of studies on electrochemical (EC) and PEC sensors for ovarian cancer biomarkers in the last seven years (2016–2022). EC sensors were included because PEC is an improved EC; and a comparison of both systems has, expectedly, been carried out in many studies. Specific attention was given to the different markers of ovarian cancer and the EC/PEC sensing platforms developed for their detection/quantification. Relevant articles were sourced from the following databases: Scopus, PubMed Central, Web of Science, Science Direct, Academic Search Complete, EBSCO, CORE, Directory of open Access Journals (DOAJ), Public Library of Science (PLOS), BioMed Central (BMC), Semantic Scholar, Research Gate, SciELO, Wiley Online Library, Elsevier and SpringerLink.

Electrochemical DNA Biosensors with Dual-Signal Amplification Strategy for Highly Sensitive HPV 16 Detection

Cervical cancer is an important topic in the study of global health issues, ranking fourth among women’s cancer cases in the world. It is one of the nine major cancers that China is focusing on preventing and treating, and it is the only cancer that can be prevented through vaccination. Systematic and effective screening for human papilloma (HPV) infection, which is closely linked to the development of cervical cancer, can reduce cervical cancer incidence and mortality. In this paper, an electrochemical sensor was designed to detect HPV 16 using dual-signal amplification. An APTES-modified glassy carbon electrode was used for improved stability. Gold nanoparticles and a chain amplification reaction were combined for signal amplification. The limit of detection (LOD) of this electrochemical sensor was 1.731 × 10−16 mol/L, and the linear response of the target detector range was from 1.0 × 10−13 mol/L to 1.0 × 10−5 mol/L (R2 = 0.99232). The test of serum sample recovery showed that it has good anti-interference, and the performance of all aspects was improved to different degrees compared with the previous research from the team. The designed sensor is centered around the principles of low cost, high sensitivity and stability, which provides new ideas for the future development of cervical cancer prevention and electrochemical biosensors.

AF-SENet: Classification of Cancer in Cervical Tissue Pathological Images Based on Fusing Deep Convolution Features

Cervical cancer is the fourth most common cancer in the world. Whole-slide images (WSIs) are an important standard for the diagnosis of cervical cancer. Missed diagnoses and misdiagnoses often occur due to the high similarity in pathological cervical images, the large number of readings, the long reading time, and the insufficient experience levels of pathologists. Existing models have insufficient feature extraction and representation capabilities, and they suffer from insufficient pathological classification. Therefore, this work first designs an image processing algorithm for data augmentation. Second, the deep convolutional features are extracted by fine-tuning pre-trained deep network models, including ResNet50 v2, DenseNet121, Inception v3, VGGNet19, and Inception-ResNet, and then local binary patterns and a histogram of the oriented gradient to extract traditional image features are used. Third, the features extracted by the fine-tuned models are serially fused according to the feature representation ability parameters and the accuracy of multiple experiments proposed in this paper, and spectral embedding is used for dimension reduction. Finally, the fused features are inputted into the Analysis of Variance-F value-Spectral Embedding Net (AF-SENet) for classification. There are four different pathological images of the dataset: normal, low-grade squamous intraepithelial lesion (LSIL), high-grade squamous intraepithelial lesion (HSIL), and cancer. The dataset is divided into a training set (90%) and a test set (10%). The serial fusion effect of the deep features extracted by Resnet50v2 and DenseNet121 (C5) is the best, with average classification accuracy reaching 95.33%, which is 1.07% higher than ResNet50 v2 and 1.05% higher than DenseNet121. The recognition ability is significantly improved, especially in LSIL, reaching 90.89%, which is 2.88% higher than ResNet50 v2 and 2.1% higher than DenseNet121. Thus, this method significantly improves the accuracy and generalization ability of pathological cervical WSI recognition by fusing deep features.

Triple Reporter Assay: A Non-Overlapping Luciferase Assay for the Measurement of Complex Macromolecular Regulation in Cancer Cells Using a New Mushroom Luciferase–Luciferin Pair

This study demonstrates the development of a humanized luciferase imaging reporter based on a recently discovered mushroom luciferase (Luz) from Neonothopanus nambi. In vitro and in vivo assessments showed that human-codon-optimized Luz (hLuz) has significantly higher activity than native Luz in various cancer cell types. The potential of hLuz in non-invasive bioluminescence imaging was demonstrated by human tumor xenografts subcutaneously and by the orthotopic lungs xenograft in immunocompromised mice. Luz enzyme or its unique 3OH-hispidin substrate was found to be non-cross-reacting with commonly used luciferase reporters such as Firefly (FLuc2), Renilla (RLuc), or nano-luciferase (NLuc). Based on this feature, a non-overlapping, multiplex luciferase assay using hLuz was envisioned to surpass the limitation of dual reporter assay. Multiplex reporter functionality was demonstrated by designing a new sensor construct to measure the NF-κB transcriptional activity using hLuz and utilized in conjunction with two available constructs, p53-NLuc and PIK3CA promoter-FLuc2. By expressing these constructs in the A2780 cell line, we unveiled a complex macromolecular regulation of high relevance in ovarian cancer. The assays performed elucidated the direct regulatory action of p53 or NF-κB on the PIK3CA promoter. However, only the multiplexed assessment revealed further complexities as stabilized p53 expression attenuates NF-κB transcriptional activity and thereby indirectly influences its regulation on the PIK3CA gene. Thus, this study suggests the importance of live cell multiplexed measurement of gene regulatory function using more than two luciferases to address more realistic situations in disease biology.

An SPRi Biosensor for Determination of the Ovarian Cancer Marker HE4 in Human Plasma

Human epididymis protein 4 (HE4) is an ovarian cancer marker. Various cut-off values of the marker in blood are recommended, depending on the method used for its determination. An alternative biosensor for HE4 determination in blood plasma has been developed. It consists of rabbit polyclonal antibody against HE4, covalently attached to a gold chip via cysteamine linker. The biosensor is used with the non-fluidic array SPRi technique. The linear range of the analytical signal response was found to be 2–120 pM, and the biosensor can be used for the determination of the HE4 marker in the plasma of both healthy subjects and ovarian cancer patients after suitable dilution with a PBS buffer. Precision (6–10%) and recovery (101.8–103.5%) were found to be acceptable, and the LOD was equal to 2 pM. The biosensor was validated by the parallel determination of a series of plasma samples from ovarian cancer patients using the Elecsys HE4 test and the developed biosensor, with a good agreement of the results (a Pearson coefficient of 0.989). An example of the diagnostic application of the developed biosensor is given—the influence of ovarian tumor resection on the level of HE4 in blood serum.

Data-Driven Cervical Cancer Prediction Model with Outlier Detection and Over-Sampling Methods

Globally, cervical cancer remains as the foremost prevailing cancer in females. Hence, it is necessary to distinguish the importance of risk factors of cervical cancer to classify potential patients. The present work proposes a cervical cancer prediction model (CCPM) that offers early prediction of cervical cancer using risk factors as inputs. The CCPM first removes outliers by using outlier detection methods such as density-based spatial clustering of applications with noise (DBSCAN) and isolation forest (iForest) and by increasing the number of cases in the dataset in a balanced way, for example, through synthetic minority over-sampling technique (SMOTE) and SMOTE with Tomek link (SMOTETomek). Finally, it employs random forest (RF) as a classifier. Thus, CCPM lies on four scenarios: (1) DBSCAN + SMOTETomek + RF, (2) DBSCAN + SMOTE+ RF, (3) iForest + SMOTETomek + RF, and (4) iForest + SMOTE + RF. A dataset of 858 potential patients was used to validate the performance of the proposed method. We found that combinations of iForest with SMOTE and iForest with SMOTETomek provided better performances than those of DBSCAN with SMOTE and DBSCAN with SMOTETomek. We also observed that RF performed the best among several popular machine learning classifiers. Furthermore, the proposed CCPM showed better accuracy than previously proposed methods for forecasting cervical cancer. In addition, a mobile application that can collect cervical cancer risk factors data and provides results from CCPM is developed for instant and proper action at the initial stage of cervical cancer.

A Model for Predicting Cervical Cancer Using Machine Learning Algorithms

A growing number of individuals and organizations are turning to machine learning (ML) and deep learning (DL) to analyze massive amounts of data and produce actionable insights. Predicting the early stages of serious illnesses using ML-based schemes, including cancer, kidney failure, and heart attacks, is becoming increasingly common in medical practice. Cervical cancer is one of the most frequent diseases among women, and early diagnosis could be a possible solution for preventing this cancer. Thus, this study presents an astute way to predict cervical cancer with ML algorithms. Research dataset, data pre-processing, predictive model selection (PMS), and pseudo-code are the four phases of the proposed research technique. The PMS section reports experiments with a range of classic machine learning methods, including decision tree (DT), logistic regression (LR), support vector machine (SVM), K-nearest neighbors algorithm (KNN), adaptive boosting, gradient boosting, random forest, and XGBoost. In terms of cervical cancer prediction, the highest classification score of 100% is achieved with random forest (RF), decision tree (DT), adaptive boosting, and gradient boosting algorithms. In contrast, 99% accuracy has been found with SVM. The computational complexity of classic machine learning techniques is computed to assess the efficacy of the models. In addition, 132 Saudi Arabian volunteers were polled as part of this study to learn their thoughts about computer-assisted cervical cancer prediction, to focus attention on the human papillomavirus (HPV).

Hybrid Loss-Constrained Lightweight Convolutional Neural Networks for Cervical Cell Classification

Artificial intelligence (AI) technologies have resulted in remarkable achievements and conferred massive benefits to computer-aided systems in medical imaging. However, the worldwide usage of AI-based automation-assisted cervical cancer screening systems is hindered by computational cost and resource limitations. Thus, a highly economical and efficient model with enhanced classification ability is much more desirable. This paper proposes a hybrid loss function with label smoothing to improve the distinguishing power of lightweight convolutional neural networks (CNNs) for cervical cell classification. The results strengthen our confidence in hybrid loss-constrained lightweight CNNs, which can achieve satisfactory accuracy with much lower computational cost for the SIPakMeD dataset. In particular, ShufflenetV2 obtained a comparable classification result (96.18% in accuracy, 96.30% in precision, 96.23% in recall, and 99.08% in specificity) with only one-seventh of the memory usage, one-sixth of the number of parameters, and one-fiftieth of total flops compared with Densenet-121 (96.79% in accuracy). GhostNet achieved an improved classification result (96.39% accuracy, 96.42% precision, 96.39% recall, and 99.09% specificity) with one-half of the memory usage, one-quarter of the number of parameters, and one-fiftieth of total flops compared with Densenet-121 (96.79% in accuracy). The proposed lightweight CNNs are likely to lead to an easily-applicable and cost-efficient automation-assisted system for cervical cancer diagnosis and prevention.

RGB Channel Superposition Algorithm with Acetowhite Mask Images in a Cervical Cancer Classification Deep Learning Model

Cervical cancer is one of the main causes of death from cancer in women. However, it can be treated successfully at an early stage. This study aims to propose an image processing algorithm based on acetowhite, which is an important criterion for diagnosing cervical cancer, to increase the accuracy of the deep learning classification model. Then, we mainly compared the performance of the model, the original image without image processing, a mask image made with acetowhite as the region of interest, and an image using the proposed algorithm. In conclusion, the deep learning classification model based on images with the proposed algorithm achieved an accuracy of 81.31%, which is approximately 9% higher than the model with original images and approximately 4% higher than the model with acetowhite mask images. Our study suggests that the proposed algorithm based on acetowhite could have a better performance than other image processing algorithms for classifying stages of cervical images.

An Advanced 3D Model of Vascularized Epithelial Ovarian Cancer in a Tumor-on-a-Chip System Based on Multi-Cell Culture

Epithelial ovarian cancer (EOC) is a highly lethal malignancy characterized by significant heterogeneity and poor prognosis due to late-stage diagnosis and chemotherapy resistance. Traditional two-dimensional (2D) models fail to replicate the complexity of the tumor microenvironment (TME), necessitating the development of advanced in vitro systems. Here, we present a novel microfluidic tumor-on-a-chip (ToC) system that accurately models key features of EOC, including heterogeneity and vascularization. The developed cellular model was evaluated for functionality. It was demonstrated that endothelial cells of blood vessels within a collagen matrix successfully migrated toward the cancerous tissue, while the multicellular and multilayered tumor construct secreted pro-angiogenic factors. Additionally, long-term culture conditions induced inflammatory responses, mimicking in vivo tumor progression. This innovative platform enables precise investigations into EOC biology, angiogenesis, and TME interactions. Furthermore, it holds significant potential for drug screening, assessing therapeutic efficacy, and advancing personalized oncology approaches.