Skip to content

LungFlag

Overview

The LungFlag model was developed at Kaiser Permanente Southern California (KPSC) to identify all types of lung cancer, including Squamous, Adenocarcinoma, Small Cell, and Non-Small Cell variants.

It has been validated at approximately 10 different sites.

Model Inputs

Inputs are listed in the /discovery API, including signal names and required units.

  • Birth year (for age calculation)
  • Sex
  • Measurements: BMI, Weight, Height
  • Smoking Information: Smoking Status (mandatory), Smoking duration (years), smoking intensity (ciggarets per day), Pack years, Smoking quit date (if applicable)
  • Diagnosis signal: ICD10/ICD9
  • Spirometry test: Fev1
  • CBC panel: Hemoglobin, Hematocrit, RBC, MCH, MCV, MCHC, Platelets, RDW, WBC, Lymphocytes (absolute and %), Monocytes (absolute and %), Eosinophils (absolute and %), Basophils (absolute and %), Neutrophils (absolute and %)
  • Lipid panel: Cholesterol, Triglycerides, LDL, HDL, NonHDLCholesterol
  • Basic Metabolic Panel (BMP): Glucose, Creatinine, Urea
  • Comprehensive Metabolic Panel (CMP) additions: Albumin, Protein_Total, ALT, ALKP

Deployments

  • Geisinger Health System (since 2022)

Intended Usage

For further information, please see the "Contact Details for Usage" section and request the User Guide.

List of Publications

  • Partial list of publications.
Manuscript Population Year
Machine Learning for Early Lung Cancer Identification Using Routine Clinical and Laboratory Data KPSC - US 2021
Flagging High-Risk Individuals with a ML Model Improves NSCLC Early Detection in a USPSTF-Eligible Population KPSC - US 2023
Improved Efficiency with Lungflag vs Opportunistic Selection in a Theoretical East Asian Lung Cancer Screening Program Budget impact 2023
Use of the predictive risk model LungFlagTM for lung cancer screening in screening in a Spanish reference center: A cost-effectiveness analysis Spain - Budget Impact 2023
LungFlag, a machine-learning (ML) personalized tool for assessing lung cancer risk in a community setting, to evaluate performance in flagging non-small cell lung cancer (NSCLC) regardless of sex or race KPSC - US 2023
1561P Targeted screening methodologies to select high risk individuals: LungFlag performance in Estonia Lung Cancer Screening Pilot Estonia 2024
Validation of LungFlagā„¢ Prediction Model Using Electronic Medical Records (EMR) On Taiwan Data National Taiwan University Hospital - Taiwan 2024
Artificial intelligenceā€“aided lung cancer screening in routine clinical practice: A pilot of LungFlag at Geisinger Geisinger - US 2024
Budget impact model of LungFlag, a predictive risk model for lung cancer screening US - Budget Impact 2024
Validation of LungFlag Lean machine-learning model to identify individuals with lung cancer using multinational data KPSC - US, Geisinger - US, THIN - UK, additional site in US 2025
Cost-effectiveness of a machine learning risk prediction model (LungFlag) in the selection of high-risk individuals for non-small cell lung cancer screening in Spain Spain - Budget Impact 2025
Maximizing Lung Cancer Screening in High-Risk Population Leveraging ML-Developed Risk-Prediction Algorithms: Danish Retrospective Validation of LungFlag Southern Denmark 2025

Posters

Poster Population Year
Flagging high-risk individuals with a ML model improves NSCLC early detection in a USPSTF-eligible population KPSC - US 2022
Computer-assisted Flagging of Never Smokers at High Risk of NSCLC in a Large US-based HMO using the LungFlag Model KPSC - US 2022
Cost-effectiveness of a machine learning risk prediction model (LungFlag ) in the selection of high-risk individuals for non-small cell lung cancer screening in Spain Budget impact 2023
Internal Poster - Real-World Evidence as the centerpiece for the evaluation of LungFlag pre-screening digital algorithm Budget Impact 2023
LungFlag, a Machine-Learning (ML) Personalized Tool for Assessing Pulmonary Complications a Community Setting, Demonstrates Comparable Performance in Flagging Non-Small Cell Lung Cancer (NSCLC) Regardless of Sex or Race KPSC - US 2023
Improved Efficiency with LungFlag vs. Opportunistic Selection in a Theoretical East Asian Lung Cancer Screening Program Budget Impact 2023
Use of the predictive risk model LungFlag for lung cancer screening in a Spanish reference center: A cost-effectiveness analysis Budget Impact 2023
Artificial intelligenceā€“aided lung cancer screening in routine clinical practice: A pilot of LungFlag at Geisinger Geisinger - US 2024
Budget impact model of LungFlag, a predictive risk model for lung cancer screening Budget Impact 2024
Validation of LungFlagā„¢ Prediction Model Using Electronic Medical Records (EMR) On Taiwan Data National Taiwan University Hospital - Taiwan 2024
TARGETED SCREENING METHODOLOGIES TO SELECT HIGH RISK INDIVIDUALS: LUNGFLAGā„¢ PERFORMANCE IN ESTONIAN LUNG CANCER SCREENING PILOT Estonia 2024
LungFlag Risk Prediction Validation on Canadian Ever Smokers Pre-Classified as High Risk for Lung Cancer Canada 2025

On Going Studies

  • Validation on CPRD
  • Validation on 4 datasets

Contact Details for Usage

Roche Navify Algosuit - Details to be announced.