Employing a Backpropagation Neural Network for Predicting Fear of Cancer Recurrence among Non-Small Cell Lung Cancer Patients

Objective:Non-small cell lung cancer(NSCLC)patients often experience significant fear of recurrence.To facilitate precise identification and appropriate management of this fear,this study aimed to compare the efficacy and accuracy of a Backpropagation Neural Network(BPNN)against logistic regression in modeling fear of cancer recurrence prediction.Methods:Data from 596 NSCLC patients,collected between September 2023 and December 2023 at the Cancer Hospital of the Chinese Academy of Medical Sciences,were analyzed.Nine clinically and statistically significant variables,identified via univariate logistic regression,were inputted into both BPNN and logistic regression models developed on a training set(N=427)and validated on an independent set(N=169).Model performances were assessed using Area Under the Receiver Operating Characteristic(ROC)Curve and Decision Curve Analysis(DCA)in both sets.Results:The BPNN model,incorporating nine selected variables,demonstrated superior performance over logistic regression in the training set(AUC=0.842 vs.0.711,p<0.001)and validation set(0.7 vs.0.675,p<0.001).Conclusion:The BPNN model outperforms logistic regression in accurately predicting fear of cancer recurrence in NSCLC patients,offering an advanced approach for fear assessment.

Network Structure and Variability of Recurrence Fear in Early-Stage Non-Small Cell Lung Cancer:A Symptom Network Analysis

Background:Lung cancer,one of the most prevalent and deadly malignancies worldwide,not only poses a significant physical burden but also a profound psychological challenge to patients.Among these psychological challenges,the fear of recurrence stands out as a particularly distressing issue.This fear,often rooted in the patients’past experiences with the disease and its treatment,can significantly impact their quality of life,mental health,and even compliance with follow-up care.Moreover,this fear can be exacerbated by the lack of understanding and support from healthcare professionals and family members,further isolating patients and compounding their psychological burden.Therefore,understanding and addressing the fear of recurrence in lung cancer patients is crucial for improving their overall well-being and outcomes.Aims:This study aims to develop a symptom network model for fear of recurrence in early-stage lung cancer patients,analyzing symptom correlations to enhance healthcare providers’understanding and management of these symptoms,thereby improving patient outcomes and quality of life.Design:A cross-sectional study design was used.Method:We employed convenience sampling to recruit 551 lung cancer patients from the Thoracic Surgery Department of a tertiary hospital in Beijing between January 2023 and December 2023.A cross-sectional study was conducted using the General Information Questionnaire,Fear of Disease Progression Scale,and Level of Hope Scale.Network analysis was performed with JASP 0.18.3.0 using the EBICglasso method,and centrality metrics including Betweenness,Closeness,Degree centrality,and Expected influence were calculated.Results:Symptom network analysis identified fear of family impact and future work disruption as central to recurrence fear in these patients.Gender-based analysis revealed‘fear of being unable to continue work’as central in males,while‘fear of affecting family members’was central in females.Among adolescents,concerns about future work,medication side effects,and family impact showed the highest expected influence.In contrast,older patients predominantly feared major treatment implications.One-way ANOVA indicated that older age correlated with reduced recurrence fear,and higher hope levels significantly mitigated this fear.Conclusion:This study broadens understanding of fear of recurrence across demographic variables like gender and age,elucidating symptom interrelations and impacts.Future strategies should focus on patient-specific differences in recurrence fear to formulate targeted interventions.Relevance to Clinical Practice:Through in-depth analysis of the symptom network,healthcare professionals can more comprehensively understand the psychological responses of lung cancer patients when they face the risk of recurrence,and then formulate more precise and personalized treatment plans.At the same time,doctors and nurses can adjust treatment strategies in a timely manner according to the changes in the patient’s symptom network and provide more comprehensive psychological support,thus enhancing the patient’s treatment adherence and outcome.Patient Contribution:People who were invited to participate voluntarily completed a range of questionnaires.