A Multi-Hospital Clinical Dataset for Lung Cancer Mortality and Survival Analysis in Dhaka, Bangladesh

This dataset contains a comprehensive collection of clinical and demographic information of 1,782 patients diagnosed with lung cancer, gathered from multiple tertiary-level hospitals and medical centers in Dhaka, Bangladesh. The dataset has been carefully designed to reflect real-world hospital-based data collection practices, ensuring clinical realism, statistical consistency, and practical usability for medical data analytics and machine learning research.

The dataset includes 17 attributes, covering demographic details, clinical history, lifestyle factors, diagnostic findings, treatment information, and survival outcomes. Key demographic features include patient age, gender, country, and hospital source. Clinical attributes capture disease stage, family cancer history, smoking status, body mass index (BMI), cholesterol levels, and comorbidities such as hypertension, asthma, cirrhosis, and prior cancer history. Diagnostic and treatment-related variables include diagnosis date, treatment modality, end-of-treatment date, and survival outcome.

The dataset is particularly suitable for lung cancer mortality analysis, survival prediction, disease progression modeling, risk factor assessment, and explainable artificial intelligence (XAI)–based clinical decision support systems. The diversity of clinical and lifestyle variables enables robust predictive modeling and interpretability-focused research, supporting the development of data-driven healthcare solutions.

To preserve patient privacy and ethical standards, all records are fully anonymized, and no personally identifiable information is included. This dataset is intended strictly for academic research, education, and non-commercial use.

𝗞𝗲𝘆𝘄𝗼𝗿𝗱𝘀: Lung cancer, mortality prediction, clinical dataset, healthcare analytics, machine learning, explainable AI, survival analysis, medical data mining, Bangladesh.

𝗗𝗮𝘁𝗮 𝗖𝗼𝗹𝗹𝗲𝗰𝘁𝗶𝗼𝗻
The dataset represents aggregated patient records collected from four major hospitals and medical centers in Dhaka, Bangladesh, including:
1. National Institute of Cancer Research & Hospital, Dhaka
2. Evercare Hospital Dhaka
3. United Hospital Limited, Dhaka
4. Square Hospital Ltd., Dhaka

The data were compiled following a structured clinical reporting format, capturing patient demographics, diagnosis details, lifestyle risk factors, comorbidities, treatment types, and outcomes.

𝗣𝗼𝘁𝗲𝗻𝘁𝗶𝗮𝗹 𝗔𝗽𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻𝘀
1. Lung cancer mortality prediction
2. Survival analysis
3. Risk stratification
4. Clinical decision support systems
5. Explainable AI (XAI)-based healthcare models
6. Epidemiological studies