A novel approach to simulate cellular neural networks (CNN) is presented in this paper. The approach, time-multiplexing simulation, is prompted by the need to simulate hardware models and test hardware implementations of CNN. For practical applications, due to hardware limitations, it is impossible to have a one-to-one mapping between the CNN hardware processors and all the pixels of the image. This simulator provides a solution by processing the input image block by block, with the number of pixels in a block being the same as the number of CNN processors in the hardware. The algorithm for implementing this simulator is presented along with popular numerical integration algorithms. Some simulation results and comparisons are also presented.

Alzheimer’s disease (AD) is a progressive neurodegenerative disease and the current diagnosis tools, which are used, including clinical examination and neuroimaging, are costly, time consuming and complicated to generalize across demographics. Traditional machine learning (ML) methods such as Support Vector Machines (SVM), LightGBM, and Random Forest (RF) are known to be having weakness in terms of feature selection, class imbalance, and generalization. This paper presents a new Enhanced TabTransformer with Self-Supervised Learning (ETT-SSL) framework of AD classification. The structure combines the transformer-based architecture, self-supervised learning (SSL) for feature representation, SHAP-based feature selection, and the use of SMOTE to balance classes. Experimental results also indicate that ETT-SSL reaches a high accuracy of 95.8%, which is a much better result than baseline models (SVM: 72.1, RF: 78.3, LightGBM: 80.5) and even the standard TabTransformer (85.2%). Additionally, ETT-SSL offers greater accuracy and improved recall, overcoming the issue of false negatives in AD diagnosis. SHAP analysis can be used to improve transparency in clinical decision-making by making use of the model interpretability. The proposed framework recommends a salivary, interpretability and clinically practical methodology that can be used to expand to multimodal data source (e.g. MRI scans, genomic markers and electronic health records) to improve accuracy and generalisability.

In this paper, we introduce QProteoML, a new quantum machine learning (QML) framework for predicting drug sensitivity in Multiple Myeloma (MM) using high-dimensional proteomic data. MM, an extremely heterogeneous condition, displays often mixed responses to treatment, with a large number of patients showing drug resistance to proteasome inhibitors and immune modulatory agents. However, the methods previously used for genomic and proteomic data analysis techniques are plagued by issues of high dimensionality, imbalanced class distribution and feature redundancy, which work against the accurate predictability and generalizability of such methods. These are compounded by the so-called “curse of dimensionality”, with dimensions far outnumbering samples, hence classical model overfitting. In this work, we present QProteoML as an integration of quantum techniques purposefully developed to deal with high-dimensional, imbalanced and redundant data. The framework integrates a combination of Quantum Support Vector Machine (QSVM), Quantum Principal Component Analysis (qPCA), Quantum Annealing (QA) for feature selection and Quantum Generative Adversarial Networks (QGANs) for data augmentation. These quantum algorithms exploit certain quantum phenomena (superposition and entanglement) to perform modelling of nonlinear relationships, dimensionality reduction, and class-imbalance issues. QSVM employs quantum kernels to map data into a higher-dimensional Hilbert space, so that the model can detect complex patterns in MM drug resistance. qPCA reduces dimensionality without loss of important variance, and thus improves computation efficiency. In addition, Quantum Annealing successfully extracts the most informative biomarkers with low redundancy. QProteoML was experimentally tested by comparing accuracy, F1 score and AUC ROC between classical machine learning models such as Support Vector Machine (SVM), Random Forest (RF), Logistic Regression (LR), and K-Nearest Neighbors (KNN). Our results demonstrate that QProteoML performs better than classical models, particularly in identifying the drug resistant minority class of patients. Additionally, the model is interpretable and stresses important biomarkers of drug sensitivity in MM. This research opens the possibility of quantum machine learning in personalised medicine for Multiple Myeloma. It demonstrates that quantum algorithms can perform complex biological data suggesting more reliable and accurate drug sensitivity predictions. Future research will be directed toward clinical validation of the given system with larger and more diverse cohorts of MM patients; the integration of quantum hardware for practical applications.

In this paper, a universal simulator for cellular neural network (CNN) is presented. This simulator is capable of performing Raster simulation for any size of input image, and thus is a powerful tool for researchers investigating potential applications of CNN. This paper reports the latency properties of CNNs along with popular numerical integration algorithms; results and comparisons are also presented.

This paper presents, a design method for the template of the hole-filler used to improve the performance of the character recognition using Numerical integration algorithms. This is done by analyzing the features of the hole-filler template and the dynamic process of CNN and by using popular numerical algorithms to obtain a set of inequalities satisfying its output characteristics as well as the parameter range of the hole-filler template. Some simulation results and comparisons are also presented.