Predicting Thalassemia Using Feature Selection Techniques: A Comparative Analysis.

Thalassemia represents one of the most common genetic disorders worldwide, characterized by defects in hemoglobin synthesis. The affected individuals suffer from malfunctioning of one or more of the four globin genes, leading to chronic hemolytic anemia, an imbalance in the hemoglobin chain ratio, iron overload, and ineffective erythropoiesis. Despite the challenges posed by this condition, recent years have witnessed significant advancements in diagnosis, therapy, and transfusion support, significantly improving the prognosis for thalassemia patients. This research empirically evaluates the efficacy of models constructed using classification methods and explores the effectiveness of relevant features that are derived using various machine-learning techniques. Five feature selection approaches, namely Chi-Square (χ2), Exploratory Factor Score (EFS), tree-based Recursive Feature Elimination (RFE), gradient-based RFE, and Linear Regression Coefficient, were employed to determine the optimal feature set. Nine classifiers, namely K-Nearest Neighbors (KNN), Decision Trees (DT), Gradient Boosting Classifier (GBC), Linear Regression (LR), AdaBoost, Extreme Gradient Boosting (XGB), Random Forest (RF), Light Gradient Boosting Machine (LGBM), and Support Vector Machine (SVM), were utilized to evaluate the performance. The χ2 method achieved accuracy, registering 91.56% precision, 91.04% recall, and 92.65% f-score when aligned with the LR classifier. Moreover, the results underscore that amalgamating over-sampling with Synthetic Minority Over-sampling Technique (SMOTE), RFE, and 10-fold cross-validation markedly elevates the detection accuracy for αT patients. Notably, the Gradient Boosting Classifier (GBC) achieves 93.46% accuracy, 93.89% recall, and 92.72% F1 score.

Engineering of Multifunctional Nanocomposite Membranes for Wastewater Treatment: Oil/Water Separation and Dye Degradation.

Multifunctional membrane technology has gained tremendous attention in wastewater treatment, including oil/water separation and photocatalytic activity. In the present study, a multifunctional composite nanofiber membrane is capable of removing dyes and separating oil from wastewater, as well as having antibacterial activity. The composite nanofiber membrane is composed of cellulose acetate (CA) filled with zinc oxide nanoparticles (ZnO NPs) in a polymer matrix and dipped into a solution of titanium dioxide nanoparticles (TiO2 NPs). Membrane characterization was performed using transmission electron microscopy (TEM), field emission scanning electron microscopy (FESEM), and Fourier transform infrared (FTIR), and water contact angle (WCA) studies were utilized to evaluate the introduced membranes. Results showed that membranes have adequate wettability for the separation process and antibacterial activity, which is beneficial for water disinfection from living organisms. A remarkable result of the membranes' analysis was that methylene blue (MB) dye removal occurred through the photocatalysis process with an efficiency of ~20%. Additionally, it exhibits a high separation efficiency of 45% for removing oil from a mixture of oil-water and water flux of 20.7 L.m-2 h-1 after 1 h. The developed membranes have multifunctional properties and are expected to provide numerous merits for treating complex wastewater.

Intracranial Hemorrhage Detection Using Parallel Deep Convolutional Models and Boosting Mechanism.

Intracranial hemorrhage (ICH) can lead to death or disability, which requires immediate action from radiologists. Due to the heavy workload, less experienced staff, and the complexity of subtle hemorrhages, a more intelligent and automated system is necessary to detect ICH. In literature, many artificial-intelligence-based methods are proposed. However, they are less accurate for ICH detection and subtype classification. Therefore, in this paper, we present a new methodology to improve the detection and subtype classification of ICH based on two parallel paths and a boosting technique. The first path employs the architecture of ResNet101-V2 to extract potential features from windowed slices, whereas Inception-V4 captures significant spatial information in the second path. Afterwards, the detection and subtype classification of ICH is performed by the light gradient boosting machine (LGBM) using the outputs of ResNet101-V2 and Inception-V4. Thus, the combined solution, known as ResNet101-V2, Inception-V4, and LGBM (Res-Inc-LGBM), is trained and tested over the brain computed tomography (CT) scans of CQ500 and Radiological Society of North America (RSNA) datasets. The experimental results state that the proposed solution efficiently obtains 97.7% accuracy, 96.5% sensitivity, and 97.4% F1 score using the RSNA dataset. Moreover, the proposed Res-Inc-LGBM outperforms the standard benchmarks for the detection and subtype classification of ICH regarding the accuracy, sensitivity, and F1 score. The results prove the significance of the proposed solution for its real-time application.

A Performance Comparison of Classification Algorithms for Rose Plants.

One of the key roles of Botanists is to be able to recognize flowers. This role has become highly challenging given that the number of discovered flower types are nearing half a million. To support Botanists, Information Technology offers promising solutions. Specifically, machine learning techniques are intrinsically appealing due to being precise enough as required. To this aim, two observations on flower leaves are relevant and leverage flower identification: one, flower plants exhibit unique features in their leaves thus allow distinction of their co-located flowers; two, leaves have a much longer life than flowers thus preserve identity properties longer. This paper proposes the use of machine learning-based identification of rose types by leveraging the features from their leaves. For this purpose, the performance of Naive Bayes, Generalized Linear Model, Multilayer Perceptron, Decision Tree, Random Forest, Gradient Boosted Trees, and Support Vector Machine has been analyzed. This study optimizes the RF model by investigating and tuning its various parameters such as the number of trees, the depth of trees, and splitting criteria. The best results are achieved with gain ratio because it takes more distinct values to avoid the problems associated with Information Gain. Optimizing the number of trees and the depth of trees of RF yield better accuracy than other models. Extensive experiments are performed to analyze the results of ensemble algorithms by using the voting method for each instance. Results suggest that the performance of ensemble classifiers is superior to that of individual models.

Recognition of Handwritten Medical Prescription Using Signature Verification Techniques.

Patient record keeping plays a vital role in diagnoses and cures. Due to a shortage of time, most doctors write prescriptions manually in Pakistan. At times, it becomes difficult for pharmacists to read prescriptions properly. As a result, they may dispense the wrong medicine. This might cause risky and deadly effects on the patient's health. This paper proposes an online handwritten medical prescription recognition system that lets doctors write prescriptions on a tablet using a stylus and automatically recognizes the medicine. We use signature verification techniques to recognize the doctor's handwriting to overcome the problem of misinterpretation of the medicine name by the pharmacist. The proposed system stores different features like the pen coordinates, time, and several pen-ups and pen-downs. Besides using features already proposed in the literature for signature verification, we propose some new features that greatly enhance recognition accuracy. We built a dataset of 24 medicine names from two users and compared results using newly proposed features. We have obtained 84%, 78%, 77.47% 77.31%, 74.17%, 60%, 38.5%, 68%, and 61.64% accuracies for 9 users using SVM classifier.

Experimental Investigations on Dry Sliding Wear Behavior of Kevlar and Natural Fiber-Reinforced Hybrid Composites through an RSM-GRA Hybrid Approach.

The present work aimed to investigate the dry sliding wear behaviors of hybrid polymer matrix composites made up of Kevlar, bamboo, palm, and Aloe vera as reinforcement materials of varying stacking sequences, along with epoxy as the matrix material. Three combinations of composite laminates with different stacking sequences such as AB, BC, and CA were fabricated by a vacuum-assisted compression molding process. The influence of composite laminates fabricated through various stacking sequences and dry sliding wear test variables such as load, sliding distance, and sliding velocity on the specific wear rate and co-efficient of friction were investigated. Experiments were designed and statistical validation was performed through response surface methodology-based D-optimal design and analysis of variance. The optimization was performed using grey relational analysis (GRA) to identify the optimal parameters to enhance the wear resistance of hybrid polymer composites under dry sliding conditions. The optimal parameters, such as composite combinations of CA, a load of 5 N, a sliding velocity of 3 m/s, and a sliding distance of 1500 m, were obtained. Furthermore, the morphologies of worn-out surfaces were investigated using SEM analysis.

Deep Learning-Based Transfer Learning for Classification of Skin Cancer.

One of the major health concerns for human society is skin cancer. When the pigments producing skin color turn carcinogenic, this disease gets contracted. A skin cancer diagnosis is a challenging process for dermatologists as many skin cancer pigments may appear similar in appearance. Hence, early detection of lesions (which form the base of skin cancer) is definitely critical and useful to completely cure the patients suffering from skin cancer. Significant progress has been made in developing automated tools for the diagnosis of skin cancer to assist dermatologists. The worldwide acceptance of artificial intelligence-supported tools has permitted usage of the enormous collection of images of lesions and benevolent sores approved by histopathology. This paper performs a comparative analysis of six different transfer learning nets for multi-class skin cancer classification by taking the HAM10000 dataset. We used replication of images of classes with low frequencies to counter the imbalance in the dataset. The transfer learning nets that were used in the analysis were VGG19, InceptionV3, InceptionResNetV2, ResNet50, Xception, and MobileNet. Results demonstrate that replication is suitable for this task, achieving high classification accuracies and F-measures with lower false negatives. It is inferred that Xception Net outperforms the rest of the transfer learning nets used for the study, with an accuracy of 90.48. It also has the highest recall, precision, and F-Measure values.

Dilated Semantic Segmentation for Breast Ultrasonic Lesion Detection Using Parallel Feature Fusion.

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.

Effective Sequestration of Phosphate and Ammonium Ions by the Bentonite/Zeolite Na-P Composite as a Simple Technique to Control the Eutrophication Phenomenon: Realistic Studies.

A bentonite/Zeolite-P (BE/ZP) composite was synthesized by controlled alkaline hydrothermal treatment of bentonite at 150 °C for 4 h for effective sequestration of phosphate and ammonium pollutants. The composite is of 512 m2/g surface area, 387 meq/100 g ion-exchange capacity, and 5.8 nm average pore diameter. The experimental investigation reflected the strong effect of the pH value in directing the uptake behavior and the best results were attained at pH 6. The kinetic properties showed an excellent agreement for phosphate and ammonium adsorption results with the pseudo-second-order model showing equilibrium intervals of 600 and 360 min, respectively, and maximum experimental capacities of 170 and 155 mg/g, respectively. Additionally, their equilibrium modeling confirmed excellent fitness with the Langmuir hypothesis, signifying homogeneous and monolayer uptake processes with a theoretical q max of 179.4 and 166 mg/g for phosphate and ammonium, respectively. Moreover, the calculated Gaussian adsorption energies of phosphate (0.8 kJ/mol) and ammonium (0.72 kJ/mol) suggested physisorption for them with mechanisms close to the zeolitic ion-exchange process or the coulumbic attractive forces. This was supported by the assessed thermodynamic parameters which also suggested spontaneous uptake by endothermic reaction for phosphate and exothermic reaction for ammonium. The BE/ZP composite is of excellent reusability and used for eight recyclability runs achieving removal percentages of 61.5 and 74.5% for phosphate and ammonium, respectively, in run 8. Finally, the composite was applied in the purification of sewage water and groundwater, achieving complete removal for phosphate from sewage water and ammonium from groundwater and reduction of the ammonium ions in the sewage water to 2.3 mg/L.

Experimental Analysis on the Influence and Optimization of µ-RUM Parameters in Machining Alumina Bioceramic.

Fabrication of precise micro-features in bioceramic materials is still a challenging task. This is because of the inherent properties of bioceramics, such as low fracture toughness, high hardness, and brittleness. This paper places an emphasis on investigating the multi-objective optimization of fabrication of microchannels in alumina (Al2O3) bioceramics by using rotary ultrasonic machining (RUM). The influence of five major input parameters, namely vibration frequency, vibration amplitude, spindle speed, depth of cut, and feed rate on the surface quality, edge chipping, and dimensional accuracy of the milled microchannels was analyzed. Surface morphology and microstructure of the machined microchannels were also evaluated and analyzed. Unlike in previous studies, the effect of vibration frequency on the surface morphology and roughness is discussed in detail. A set of designed experiments based on central composite design (CCD) method was carried out. Main effect plots and surface plots were analyzed to detect the significance of RUM input parameters on the outputs. Later, a multi-objective genetic algorithm (MOGA) was employed to determine the optimal parametric conditions for minimizing the surface roughness, edge chipping, and dimensional errors of the machined microchannels. The optimized values of the surface roughness (Ra and Rt), side edge chipping (SEC), bed edge chipping (BEC), depth error (DE), and width error (WE) achieved through the multi-objective optimization were 0.27 µm, 2.7 µm, 8.7 µm, 8 µm, 5%, and 5.2%, respectively.