Field Of Medical Diagnosis Research Articles

Applications of Gelatin in Biosensors: Recent Trends and Progress.

Gelatin is a natural protein from animal tissue with excellent biocompatibility, biodegradability, biosafety, low cost, and sol–gel property. By taking advantage of these properties, gelatin is considered to be an ideal component for the fabrication of biosensors. In recent years, biosensors with gelatin have been widely used for detecting various analytes, such as glucose, hydrogen peroxide, urea, amino acids, and pesticides, in the fields of medical diagnosis, food testing, and environmental monitoring. This perspective is an overview of the most recent trends and progress in the development of gelatin-based biosensors, which are classified by the function of gelatin as a matrix for immobilized biorecognition materials or as a biorecognition material for detecting target analytes.

Review on fuzzy multi-criteria decision-making methods in medical diagnosis

In a real-world computing environment, the information is not complete, precise, and certain, making it very difficult to derive an actual decision. Uncertainty is a factor that makes the decision-making process more difficult in medical diagnosis. Everyone is different from the other at the physical level as well as mental level, symptoms of a disease expressed by patients are linguistic in nature and vary from person to person. That is why Decision making in medical diagnosis is becoming a vast research field in medical science. In medical diagnosis, generally, uncertainty (vagueness or imprecision) arises due to patients’ vague linguistic expression of their problems to medical experts. The field of medicine and decision-making are the most fruitful and interesting areas of applications of fuzzy set theory. In the present paper, we study the importance of different fuzzy sets and fuzzy multi-criteria decision-making methods in the field of medical diagnosis.

Acquition of ECG and Iris data for biometric authentication

Use of Electrocardiographic signals is not new in the field of medical diagnosis to diagnose one’s heart health since long decades. But now the same signals can also be used in recognizing a person and thus as a biometric trait. Acquisition of such a vital signal can be done through wired and wireless contacts of electrodes. Several works are already being carried out in these fields. This is because of the universality, individuality, permanence, recordability of the signal from the subject. The data can be collected from the subject using single lead, three leads or even 12 leads. The ECG can be also be recorded through the subject’s fingertip using one lead ECG setup. However these signals will be vulnerable to numerous types of noises like base line wander, powerline interference, high frequency noise and low frequency noise. Thus it’s a challenging factor to design a biometric system by eradicating these noises. This paper presents a data acquisition system using ECG as well as IRIS where IRIS also provides unique information about individuals.

Chest X-Ray Examiner

Abstract: The most common image taken in the medical field for diagnosis of any ailment affecting the chest is chest radiography (Chest X-ray). The most common image taken in the medical field for diagnosis of any ailment affecting the chest or neighboring area is chest radiography (Chest X-ray). The use of this approach has been limited due to a scarcity of qualified radiologists. To address this issue, we are developing a computer-aided diagnosis system for chest X-ray disease classification that employs DNN (Deep Neural Network) Transfer learning. CXE (Chest X-Ray Examiner) is a web-based program that works in conjunction with our machine learning console. The most common image taken in the medical field for diagnosis of any ailment affecting the chest is chest radiography (Chest X-ray). There is a Rest API application that serves as a middleman between our user interface and machine learning applications. We used our machine learning console application produced by ML.NET to train our own model using the Mobile.Net v3 Image categorization method. By obtaining X-Ray images from end users, the Chest X-Ray Examiner (CXE) can classify the chest disease name and forecast the accuracy level of that disease. Keywords: Chest X-ray, DNN, API, ML.NET, Chest X-Ray Examiner

Design of the Floating Hologram Method with a Reverse Pyramid Type for CT and MR Diagnosis in Clinical Room

In the field of medical diagnosis, big data and three-dimensional (3D) imaging diagnosis technology are being applied due to the development of these technologies. Using radiology diagnosis methods, medical staff are increasing their understanding and ability to explain symptoms to patients, but they are experiencing difficulties due to communication problems. Therefore, if the medical staff shows the lesion by providing the patient with a 3D image, the understanding of the patient can be increased. This paper proposes the design of a system to produce an inverted pyramid-shaped floating holographic image to increase the patient’s understanding. The hologram system consists of an optical source generator and a beam mirror and utilizes a technology to plot an image using a 45° refraction angle of the beam of the optical source. Selected objects for observation were liver, colon, and lung, and to observe these tissues, a Computed Tomography (CT) image was input to the hologram system through the picture archiving and communication system (PACS), and the image was displayed. Tissues observed through the mirror can be observed from the left, right, front, and back with a 360° anterior view. Therefore, it is possible to observe at the desired position by the medical staff and the patient in the treatment room, and the image is large and clear, so it is very satisfying to observe. As a holographic imaging diagnostic system, it is expected that this study can be used in clinics, medical education rooms, and operating rooms in the future.

An efficient rotational direction heap-based optimization with orthogonal structure for medical diagnosis

The heap-based optimizer (HBO) is an optimization method proposed in recent years that may face local stagnation problems and show slow convergence speed due to the lack of detailed analysis of optimal solutions and a comprehensive search. Therefore, to mitigate these drawbacks and strengthen the performance of the algorithm in the field of medical diagnosis, a new MGOHBO method is proposed by introducing the modified Rosenbrock's rotational direction method (MRM), an operator from the grey wolf optimizer (GWM), and an orthogonal learning strategy (OL). The MGOHBO is compared with eleven famous and improved optimizers on IEEE CEC 2017. The results on benchmark functions indicate that the boosted MGOHBO has several significant advantages in terms of convergence accuracy and speed of the process. Additionally, this article analyzed the diversity and balance of MGOHBO in detail. Finally, the proposed MGOHBO algorithm is utilized to optimize the kernel extreme learning machines (KELM), and a new MGOHBO-KELM is proposed. To validate the performance of MGOHBO-KELM, seven disease diagnostic questions were introduced for testing in this work. In contrast to advanced models such as HBO-KELM and BP, it can be concluded that the MGOHBO-KELM model can achieve optimal results, which also proves that it has practical significance in solving medical diagnosis problems.

Hybrid multiagent based adaptive genetic algorithm for limited view tomography using oppositional learning

Computed tomography (CT) plays a crucial role in the field of medical diagnosis. The prime objective of limited view tomography is to estimate the object’s internal structure with limited view projection data. Limited view computed tomography (CT) has the ability to reduce the X-ray radiation dose imposed on the patient. This manuscript presents Hybrid Multiagent based Adaptive Genetic Algorithm (HMAGA) for limited view tomography. HMAGA is a combination of multiagent-based genetic algorithms and simulated annealing with adaptive crossover and mutation rate. The proposed algorithm uses two methods for reducing the loss of diversity and increases the convergence rate, and these methods are oppositional learning and a new random population. Experimental results reveal that the proposed algorithm provides satisfactory results with low computation overhead. The present manuscript also outperforms other states of the art reconstruction algorithms.

Cesium Lead Halide Nanocrystals based Flexible X‐Ray Imaging Screen and Visible Dose Rate Indication on Paper Substrate

AbstractMetal halide semiconductors have huge potential on X‐ray detection, due to their high atomic number, adjustable photoluminescence wavelength, superior stability, and simple fabrication process. Here, cesium lead halide nanocrystals (CsPbX3 NCs) are fabricated on paper substrate as scintillator screen for the first time. By means of all‐solution processed spray‐coating method with optimization of polymethyl methacrylate, large‐area (>600 cm2) and low‐cost (0.0975 dollar cm−2) flexible scintillator screens are obtained with high resolution (≈5 lp mm−1), low detection limit (2.97 μGy s−1), high light yield (≈30000 photons/MeV), and fast response speed (<15 ms). The screen can achieve high‐resolution X‐ray non‐distortion imaging for non‐planar objects and realize battery check and safety monitoring at a broad range of photon energy (10–160 keV). In addition, real‐time dose rates can be easily estimated through counting the luminous arcs of the designed visible indicator. More important, the screen shows remarkable tolerance to high‐dose X‐ray radiation (tolerance over 60 Gyair continuous radiation with only 3% performance degradation), water soaking (2000 h), and conventional storage (over 300 days without apparent deterioration under ambient conditions). Hence, this work may provide huge potential in the field of medical diagnosis, non‐planar detection, and safety monitoring in the future.

Sensing absorptive fluids with backside illuminated grating coupled SPR sensor fabricated by nanoimprint technology

Grating coupled surface plasmon resonance (GC-SPR) sensors are widely used in analyzing gases, aqueous solutions, proteins and environmental pollutants. However, traditional GC-SPR sensors are difficult to detect absorptive fluids because the spectra of the light will be distorted after passing through the absorptive fluids. Here, we propose and experimentally demonstrate a backside illuminated GC-SPR (BIGC-SPR) sensor, where the light is incident from the transparent substrate side of the sensor chip. The grating chip is fabricated by a simple and cost-effective procedure. DVD-R disc is used as a template to imprint the grating structure onto a polydimethylsiloxane (PDMS) substrate, then a thin Au layer of 35 nm thickness is evaporated on it. By calibrating the sensor with glucose solutions, the sensitivity and refractive index resolution are measured to be 311.97 nm/RIU and 0.00043 RIU, respectively. Moreover, the BIGC-SPR sensor is demonstrated to monitor the concentration of blue engine coolant, green antirust cutting fluid and white antirust emulsified oil solutions. The designed sensor is able to detect colored or highly scattering solutions, and can be potentially used in the fields of medical diagnosis, environmental monitoring and biological protein detection.

Deep Learning-Based Imbalanced Classification With Fuzzy Support Vector Machine.

Imbalanced classification is widespread in the fields of medical diagnosis, biomedicine, smart city and Internet of Things. The imbalance of data distribution makes traditional classification methods more biased towards majority classes and ignores the importance of minority class. It makes the traditional classification methods ineffective in imbalanced classification. In this paper, a novel imbalance classification method based on deep learning and fuzzy support vector machine is proposed and named as DFSVM. DFSVM first uses a deep neural network to obtain an embedding representation of the data. This deep neural network is trained by using triplet loss to enhance similarities within classes and differences between classes. To alleviate the effects of imbalanced data distribution, oversampling is performed in the embedding space of the data. In this paper, we use an oversampling method based on feature and center distance, which can obtain more diverse new samples and prevent overfitting. To enhance the impact of minority class, we use a fuzzy support vector machine (FSVM) based on cost-sensitive learning as the final classifier. FSVM assigns a higher misclassification cost to minority class samples to improve the classification quality. Experiments were performed on multiple biological datasets and real-world datasets. The experimental results show that DFSVM has achieved promising classification performance.

Shared-cathode closed bipolar electrochemiluminescence cloth-based chip for multiplex detection

Electrochemiluminescence (ECL) chips have been widely used in the field of medical diagnosis. However, most of these chips currently in use are costly and require high amounts of sample. In this work, we present, for the first time, a shared-cathode closed bipolar electrochemiluminescence (SC-CBP-ECL) cloth-based chip, which can be used for multiplex detection. The SC-CBP-ECL chips ($0.03–0.05 for each chip) are manufactured using carbon ink- and wax-based screen-printing techniques, without the need for expensive and complex fabrication equipment. Under optimised conditions, the SC-CBP-ECL chips were successfully used for coinstantaneous detection of glucose in double ECL systems (i.e., Ru(bpy)32+ and luminol), with corresponding linear ranges of 0.05–1 mM and 0.05–10 mM, and detection limits of 0.0382 mM and 0.0422 mM. To our knowledge, this is the first report on the application of fibre material-based closed bipolar electrodes (C-BPE) combined with double ECL systems. Furthermore, the SC-CBP-ECL chips exhibit an acceptable specificity and good reproducibility and stability and can be used for glucose detection in human serum samples with a good agreement compared with the clinical method. Finally, the SC-CBP-ECL chips could be successfully used for simultaneous detection of seven glucose samples and also show potential for simultaneous detection of three different targets (hydrogen peroxide [H2O2], glucose, and uric acid [UA]). Therefore, we believe that the chip described in this study has broad potential application in the field of cost-effective multiplex detection.

A method for purifying nanoparticles using cationic modified monoliths and aqueous elution

Nanoparticles are widely used in the medical field for diagnosis and therapy. In particular, the use of nanoparticles containing vaccines has spread rapidly; hence, ensuring nanoparticle safety and minimizing their side effects have become important concerns worldwide. In this study, we used three types (NH2, poly-Lys, and trimethylaminopropyl) of cationic modified silica monoliths with cylindrical structures, diameters of 4.2 mm, and heights of 1.5 mm. Doxil, an anticancer nanomedicine, and exosomes, as typical nanoparticles, were separated from model leaked drugs (e.g., doxorubicin and oligonucleotides) and proteins (e.g., albumin) coexisting in nanoparticle sample solutions using these monoliths. Each nanoparticle solution (200 μL) was applied to each monolith followed by centrifugation at 9,100 g for 1 min. The ionic concentration of the elution solution was increased stepwise to determine the concentration required to elute the nanoparticles from each monolith by centrifugation. The NH2- and poly-Lys-modified monoliths separated and purified nanoparticles from leaked drugs or proteins coexisting in nanoparticle sample solutions. The nanoparticles were separated from other substances by changing the pH and concentration of the aqueous Tris buffer used as the eluent. Doxil was eluted with 500–1,000 mM Tris buffer (pH 8) when using the NH2-modified monolith, and with 200–1,000 mM Tris buffer (pH 6) when using the poly-Lys-modified monolith. Exosome was obtained using 1,000 mM Tris buffer (pH 8) and the NH2-modified monolith. The recovery efficiencies (ratio of nanoparticle content in the most abundant fraction to that in the sample solution before purification) of Doxil and exosome were 64% and 55%, respectively. Because this method can purify nanoparticles using only low-speed centrifugation for a few minutes, we expect it will be used to improve nanoparticle safety.

STBi-YOLO: A Real-Time Object Detection Method for Lung Nodule Recognition

Lung cancer is the most prevalent and deadly oncological disease in the world, but a timely detection of lung nodules can greatly improve the survival rate of this disease. However, due to the tiny size of lung nodules and inconspicuous edges, lung nodules are not easily distinguished by naked eyes thus medical image diagnosticians are prone to misdiagnosis simply based on their own experiences and subjective judgements. In recent years, the machine-learning-based image processing techniques find their wide applications in the field of medical diagnosis, and have been proved to be an efficient way to aid diagnosticians to accurately identify subtle lesions in images. To accurately recognize lung nodules in CT images, in this paper, we propose an approach, called STBi-YOLO. This approach stems from YOLO-v5, but makes significant improvements from three dimensions—we first use the spatial pyramid pooling network that is based on stochastic-pooling method to modify the basic network structure of YOLO-v5; then apply a bidirectional feature pyramid network to perform multi-scale feature fusion; finally improve the loss function of the YOLO-v5 and adopt the EIoU function to optimize the training model. To evaluate our approach, we compare STBi-YOLO with YOLO-v3, YOLO-v4, YOLO-v5, and multiple leading object detection models, such as Faster R-CNN and SSD. The experiments show that STBi-YOLO achieves an accuracy of 96.1% and a recall rate of 93.3% for the detection of lung nodules, while producing a 4× smaller model size in memory consumption than YOLO-v5 and exhibiting comparable results in terms of mAP and time cost against Faster R-CNN and SSD.

A Novel Hybridized Feature Extraction Approach for Lung Nodule Classification Based on Transfer Learning Technique.

Purpose:In the field of medical diagnosis, deep learning-based computer-aided detection of diseases will reduce the burden of physicians in the diagnosis of diseases especially in the case of lung cancer nodule classification.Materials and Methods:A hybridized model which integrates deep features from Residual Neural Network using transfer learning and handcrafted features from the histogram of oriented gradients feature descriptor is proposed to classify the lung nodules as benign or malignant. The intrinsic convolutional neural network (CNN) features have been incorporated and they can resolve the drawbacks of handcrafted features that do not completely reflect the specific characteristics of a nodule. In the meantime, they also reduce the need for a large-scale annotated dataset for CNNs. For classifying malignant nodules and benign nodules, radial basis function support vector machine is used. The proposed hybridized model is evaluated on the LIDC-IDRI dataset.Results:It has achieved an accuracy of 97.53%, sensitivity of 98.62%, specificity of 96.88%, precision of 95.04%, F1 score of 0.9679, false-positive rate of 3.117%, and false-negative rate of 1.38% and has been compared with other state of the art techniques.Conclusions:The performance of the proposed hybridized feature-based classification technique is better than the deep features-based classification technique in lung nodule classification.

Label-free detection of ssDNA base insertion and deletion mutations by surface-enhanced Raman spectroscopy.

Surface-enhanced Raman spectroscopy (SERS), as a label-free, highly sensitive analytical method, has become an important tool for providing substance fingerprints. In this study, silver nanoparticles containing thiosulfate ions and calcium ions (Ag@SCNPs) have been used as an enhanced substrate to eliminate the interference of impurities on DNA signals. Intrinsic structural information on single-strand DNA (ssDNA) was directly obtained through SERS. The improved enhancement system was used to explore the base-stacking rules of ssDNA in a solution environment. For the first time, single-base insertion mutations and deletion mutations, as well as their exact mutation sites, were identified, and Raman spectra with high stability, repeatability, and high signal-to-noise ratio were obtained. The method is simple, fast, and accurate, and the detection process is nondestructive. It has potential to be applied in the fields of medical diagnosis and genetics research.

Experimental study on micro-grinding and flow characteristics of quartz glass with micro-channel

As an important part of the microfluidic system, the microfluidic channel is a key factor to improve the microfluidic performance. The glass microfluidic channel has huge application potential in the fields of medical diagnosis, biochemical analysis and microelectromechanical systems. The micro-channel was ground on the surface of quartz glass by micro-grinding machining. The micro-ground micro-channel with different depths ranged from 80 μm to 200 μm were obtained. The influences of the machining depth and liquid flow rate on flow characteristics of micro-channel were investigated. The experimental results show that when the flow velocity was less than 2.5 m/s, the flow characteristics of the micro-channel first decreased and then increased with the increasing machining depth. When the flow velocity was greater than 2.5 m/s, it decreased as the increased depth.

Knee Implant Identification by Fine-Tuning Deep Learning Models.

Identification of implant model from primary knee arthroplasty in pre-op planning of revision surgery is a challenging task with added delay. The direct impact of this inability to identify the implants in time leads to the increase in complexity in surgery. Deep learning in the medical field for diagnosis has shown promising results in getting better with every iteration. This study aims to find an optimal solution for the problem of identification of make and model of knee arthroplasty prosthesis using automated deep learning models. Deep learning algorithms were used to classify knee arthroplasty implant models. The training, validation and test comprised of 1078 radiographs with a total of 6 knee arthroplasty implant models with anterior-posterior (AP) and lateral views. The performance of the model was calculated using accuracy, sensitivity, and area under the receiver-operating characteristic curve (AUC), which were compared against multiple models trained for comparative in-depth analysis with saliency maps for visualization. After training for a total of 30 epochs on all 6 models, the model performing the best obtained an accuracy of 96.38%, the sensitivity of 97.2% and AUC of 0.985 on an external testing dataset consisting of 162 radiographs. The best performing model correctly and uniquely identified the implants which could be visualized using saliency maps. Deep learning models can be used to differentiate between 6 knee arthroplasty implant models. Saliency maps give us a better understanding of which regions the model is focusing on while predicting the results.

Filter‐free self‐power CdSe/Sb2(S1−x,Sex)3 nearinfrared narrowband detection and imaging

AbstractAccurate and clear bioimaging is crucial in the field of medical diagnosis. High‐quality bioimaging requires to avoid the effects of ambient light as well as the absorption of biological tissues. Nearinfrared (NIR) narrowband detectors located at wavelength from 650 to 900 nm can meet these requirements; thus, they are the potential solution. In this work, we construct a filter‐free and self‐power NIR narrowband photodetector based on the structure of n‐CdSe/p‐Sb2(S1‐x,Sex)3 heterojunction, and achieve a narrow spectral response at 735 nm with a full width at half‐maximum of 35.3 nm in the detector. Further, the imaging characteristics of the NIR narrowband detector are explored, verifying the ability to narrowband detection and imaging. This filter‐free and self‐power NIR narrowband detector shows considerable promise in real‐life applications.image

A FRET-based detection of N-acetylneuraminic acid using CdSe/ZnS quantum dot and exonuclease III-assisted recycling amplification strategy

N-acetylneuraminic acid (Neu5Ac) is widely spread in many biologically significant glycans of mammals, commonly as a terminal α-glycoside. It is of great significance to develop analytical techniques for detection of Neu5Ac. Herein, a high-sensitive fluorescent biosensor for Neu5Ac has been developed based on FRET between CdSe/ZnS quantum dots (QDs) and BHQ2, as well as exonuclease III (Exo III)-assisted recycling amplification strategy. Employing the specially designed three-level FRET systems and fluorescent signal recovery mechanism, together with five-step recycling signal amplification chain reactions, an ultralow detection limit of 24 fM was achieved. Meanwhile, good linear response ranges within 0.2–12.5 pM and 12.5–1000 pM were founded. The assay has excellent performance in real sample detection, and thus offers great potential for detection of sialic acids modified glycans/lipids in the fields of medical diagnosis and food testing.

Multicolor Coding Up-Conversion Nanoplatform for Rapid Screening of Multiple Foodborne Pathogens.

Technologies for rapid screening of multiple foodborne pathogens have been urgently needed because of the complex food matrix and high outbreaks of foodborne diseases. In this study, multicolor coding up-conversion nanoparticles (UCNPs) were synthesized and applied for rapid and simultaneous detection of five kinds of foodborne pathogens. The multicolor coding UCNPs were obtained through doping different concentrations of a sensitizer (Yb3+) on the shell of the synthesized NaYF4:Yb3+, Tm3+ (20%/2%)@NaYF4:Yb3+, and Er3+ (x %/2%) core/shell nanocrystals. All the UCNPs could emit red and green luminescence simultaneously once excited with near-infrared wavelength (980 nm), and the ratio of red and green (R/G ratio) emission light intensity of each kind of UCNPs varied depending on the Yb3+ doping concentration. In addition, the magnetic nanoparticles (MNPs) modified with the monoclonal antibodies (mAbs) against the target bacteria were used to capture and separate the bacteria, resulting in obtaining the MNP-bacterium complexes. Different UCNPs with multicolor coding acted as signal probes were also modified with the mAbs to react with the MNP-bacterium complexes to form the MNP-bacterium-UCNP sandwich complexes. After the sandwich complexes were excited with a wavelength of 980 nm, the obtained R/G ratios and the green photoluminescence intensity (PL intensity) could be used to distinguish and quantitatively detect foodborne pathogens, respectively. This proposed nanoplatform could detect five foodborne pathogens simultaneously within 2 h with good sensitivity and specificity, showing great potential for multiplex detection of other targets in the fields of medical diagnosis and food security.