A bipolar fuzzy decision-making system for assessing high-risk coexisting tuberculosis disease in pregnant women

Tuberculosis (TB) diagnosis poses a formidable challenge in global healthcare, particularly impacting older individuals and pregnant women. Diagnosing TB disease during pregnancy and in comorbid patients is more challenging due to overlapping symptoms with normal pregnancy conditions and existing treatments for other diseases, necessitating careful assessment to differentiate TB symptoms from those of other underlying conditions. To address this issue, this study designs a novel bipolar fuzzy decision-support system by integrating the concept of complex proportional assessment (COPRAS) and a technique for order preference by similarity to the ideal solution (TOPSIS) approaches using bipolar heptagonal fuzzy numbers. The approach is utilized to assess the high-risk of TB coinfection disease in pregnant women. The bipolar fuzzy set provides positive and negative membership degrees of an element, which divulge a balanced perspective by both the presence and absence of the disease. Additionally, a defuzzification algorithm is proposed for bipolar heptagonal fuzzy numbers, converting bipolar heptagonal fuzzy into a bipolar crisp score (CBHpFBCS). The bipolar fuzzy entropy measure is utilized to weight the criteria. The findings highlight that TB+HIV $$ coinfection is more severe in pregnant women compared to other TB comorbidities. Finally, sensitivity and comparative analyses are executed across diverse criteria weight scenarios and with existing fuzzy multi-criteria decision-making (MCDM) methods to validate the robustness of the proposed method and its outcomes.