A New Fuzzy Modeling Framework for Integrated Risk Prognosis and Therapy of Bladder Cancer Patients

Abstract

This paper presents a new fuzzy modelling approach for analysing censored survival data and finding risk groups of patients diagnosed with bladder cancer. The proposed framework involves a new procedure for integrating the frameworks of interval type-2 fuzzy logic and Cox modelling intrinsically. The output of this synergistic framework is a score/prognostics index which is indicative of the patient's level of mortality risk. A threshold value is selected whereby patients with risk scores that are greater than this threshold are classed as high risk patients and vice versa.Unlike in the case of black-box type modelling approaches, the paper shows that interpretability and transparency are maintained using the proposed fuzzy modelling framework.Two data sets are used to test the modelling accuracy of the elicited models. The first is an artificial data set which has similar characteristics as in a typical survival data. The second relates to real-life bladder cancer data from which one requires a model that identifies the low risk and high risk patients and then recommends risk management decisions based on, predicted risk level, patient history and characteristics, disease pathology and event times. The performance of the proposed framework is compared with the traditional Cox model, logistic regression as well as a non-linear survival data modelling technique based on neural networks. This is the first time an attempt has been made to exploit the transparency advantages of fuzzy models and the principled statistical framework of the Cox model in order to identify risk groups and recommend risk management decisions from complex survival data sets. In both the artificial data and real data, the proposed modelling framework, although minimalistic, shows better generalisation performances than the previously reported models against which the results were compared.O. Obajemu and M. Mahfouf are with the

Construction and application of urinary system model with functional bladder module

“…Bladder, as an organ of the urinary system in the body, is a necessary path for urine excretion. When bladder lesions occur, such as nearby ureteral obstruction, internal cancer and other conditions that make the urinary system unable to work properly, the construction of the model is indispensable (Obajemu et al, 2018, Xiao et al, 2017, Sole et al, 2019, Oliveira et al, 2018). In urinary clinical diagnosis, patients are often required to insert the catheter for catheterization, so modeling is one of the methods to accelerate the understanding of its principle.…”

Big Medical Data Decision-Making Intelligent System Exploiting Fuzzy Inference Logic for Prostate Cancer in Developing Countries

“…The use of fuzzy inference logic for disease detection and diagnosis has become increasing popular. However, most existing disease detection methods based on fuzzy inference logic present a lack of sufficient medical data information and do not assign reasonable weights to evaluation indexes [24], [31], [32] Specifically, Caifeng and Deng [24] proposed an innovative intelligent diagnosis system using a fuzzy concept lattice. In the system, the process of retrieving medical information and diagnosing disease is performed using fuzzy concept lattices based on rules of clinical diagnosis.…”

“…Additionally, an efficient optimization strategy in the system is formed based on the first derivate information and the gradient descent method. Obajemu et al [32] recommended a novel fuzzy modeling framework for risk grouping of the patients with bladder cancer by collecting and assessing the censored survival database. Based on the type-2 fuzzy inference modeling method and tow databases of bladder cancer patients (real life and manually produced datasbases), the risk score and prognostic indicators of the disease can be precisely predicted, and then the system automatically provides doctors with therapeutic recommendations and effective risk management of the disease at different stages.…”

On an Affordable Approach towards the Diagnosis and Care for Prostate Cancer Patients Using Urine, FTIR and Prediction Machines

“…The use of computational methods within clinical medicine has seen a rapid rise in its applications over recent years, largely due to the availability of greater computational power and greater awareness of what these approaches can offer clinical settings in terms of improved disease diagnosis capability, while also allowing for an enhancement in care strategies that can be offered to patients [ 6 , 7 ]. In particular, machine-learning methods have been deeply researched in the area of cancer for such tasks as prediction of the presence of cancerous cells, survival analysis, and even treatment scheduling and optimal dosing of treatment therapies for patients [ 8 , 9 ]. With the emphasis on prostate cancer, machine-learning methods have been employed mostly towards the binary prediction of the presence of cancers, where the primary sources of data include pathological samples, radiological instrumentation, and blood samples [ 10 , 11 , 12 , 13 , 14 , 15 ].…”