An Enhanced Hyper-Parameter Optimization of a Convolutional Neural Network Model for Leukemia Cancer Diagnosis in a Smart Healthcare System.

Healthcare systems in recent times have witnessed timely diagnoses with a high level of accuracy. Internet of Medical Things (IoMT)-enabled deep learning (DL) models have been used to support medical diagnostics in real time, thus resolving the issue of late-stage diagnosis of various diseases and increasing performance accuracy. The current approach for the diagnosis of leukemia uses traditional procedures, and in most cases, fails in the initial period. Hence, several patients suffering from cancer have died prematurely due to the late discovery of cancerous cells in blood tissue. Therefore, this study proposes an IoMT-enabled convolutional neural network (CNN) model to detect malignant and benign cancer cells in the patient's blood tissue. In particular, the hyper-parameter optimization through radial basis function and dynamic coordinate search (HORD) optimization algorithm was used to search for optimal values of CNN hyper-parameters. Utilizing the HORD algorithm significantly increased the effectiveness of finding the best solution for the CNN model by searching multidimensional hyper-parameters. This implies that the HORD method successfully found the values of hyper-parameters for precise leukemia features. Additionally, the HORD method increased the performance of the model by optimizing and searching for the best set of hyper-parameters for the CNN model. Leukemia datasets were used to evaluate the performance of the proposed model using standard performance indicators. The proposed model revealed significant classification accuracy compared to other state-of-the-art models.

A neuro-fuzzy security risk assessment system for software development life cycle.

This study aims to protect software development by creating a Software Risk Assessment (SRA) model for each phase of the Software Development Life Cycle (SDLC) using an Adaptive Neuro-Fuzzy Inference System (ANFIS) model. Software developers discovered and validated the risk variables affecting each SDLC phase, following which relevant data about risk factors and associated SRA for each SDLC phase were collected. To create the SRA model for SDLC phases, risk factors were used as inputs, and SRA was used as an output. The formulated model was simulated using 70 % and 80 % of the data for training, while 30 % and 20 % were used for testing the model. The performance of the SRA models using the test datasets was evaluated based on accuracy. According to the study findings, many risk variables were discovered and confirmed for the requirement, design, implementation, integration, and operation phases of SDLC 11, 8, 9, 4, and 6, respectively. The SRA model was formulated using the risk factors using 2048, 256, 512, 16, and 64 inference rules for the requirement, design, implementation, integration, and operation phases, respectively. The study concluded that using the SRA model to assess security risk at each SDLC phase provided a secured software development process.

Corrigendum to "A decision support system for multi-target disease diagnosis: A bioinformatics approach"[Heliyon 6 (3), (March 2020) Article e03657].

[This corrects the article DOI: 10.1016/j.heliyon.2020.e03657.].

A decision support system for multi-target disease diagnosis: A bioinformatics approach.

Malaria and typhoid fever are revered for their ability to individually or jointly cause high mortality rate. Both malaria and typhoid fever have similar symptoms and are famous for their co-existence in the human body, hence, causes problem of under-diagnosis when doctors tries to determine the exact disease out of the two diseases. This paper proposes a Bioinformatics Based Decision Support System (BBDSS) for malaria, typhoid and malaria typhoid diagnosis. The system is a hybrid of expert system and global alignment with constant penalty. The architecture of the proposed system takes input diagnosis sequence and benchmark diagnosis sequences through the browser, store these diagnosis sequences in the Knowledge base and set up the IF-THEN rules guiding the diagnosis decisions for malaria, typhoid and malaria typhoid respectively. The matching engine component of the system receives as input the input sequence and applies global alignment technique with constant penalty for the matching between the input sequence and the three benchmark sequences in turns. The global alignment technique with constant penalty applies its pre-defined process to generate optimal alignment and determine the disease condition of the patient through alignment scores comparison for the three benchmark diagnosis sequences. In order to evaluate the proposed system, ANOVA was used to compare the means of the three independent groups (malaria, typhoid and malaria typhoid) to determine whether there is statistical evidence that the associated values on the diagnosis variables means are significantly different. The ANOVA results indicated that the mean of the values on diagnosis variables is significantly different for at least one of the disease status groups. Similarly, multiple comparisons tests was further used to explicitly identify which means were different from one another. The multiple comparisons results showed that there is a statistically significant difference in the values on the diagnosis variables to diagnose the disease conditions between the groups of malaria and malaria typhoid. Conversely, there were no differences between the groups of malaria and typhoid fever as well as between the groups of typhoid fever and malaria typhoid. In order to show mean difference in the diagnosis scores between the orthodox and the proposed diagnosis system, t-test statistics was used. The results of the t-test statistics indicates that the mean values of diagnosis from the orthodox system differ from those of the proposed system. Finally, the evaluation of the proposed diagnosis system is most efficient at providing diagnosis for malaria and malaria typhoid at 97% accuracy.