AI Methodologies

How Intelligent is Your AI?

ConcertAI’s team of senior machine learning engineers continue to develop advanced rules-based and predictive AI technologies for healthcare. See their latest research using AI for clinical research and real-world evidence.

Deep Learning Model Using NLP to Identify Metastatic Status from Unstructured Notes

This study was conducted to determine if contextual understanding of metastatatic status can be extracted automatically from physician notes to identify patients with metastatic breast cancer.

Read More

Snapshot of methodology to train deep learning models

AI-based assessment of progression

AI Model to Predict Slow Progressors

There are ongoing efforts to understand and predict exceptional response to existing cancer therapies, but few clinical characteristics of these patients are known. We trained a machine learning model using the ConcertAI database of oncology EMR data that includes clinical data from CancerLinQ Discovery to predict slow progression, a proxy for exceptional response, in aNSCLC in the second line setting.

Read More

Dynamic Model to Predict Metastatic Recurrence

Machine Learning models that can dynamically predict risk of metastatic breast cancer (mBC) based on cumulative historical clinical data could help guide patient care and monitoring decisions. This study looked at an ML model that sought to predict risk of recurrence at any point in the patient journey.

Read More

Depiction of a patient journey and features used to understand risk of recurrence

ML Model to Impute Cancer Subtype

Ability to distinguish between subtypes of lung cancer (LC) is important for clinical outcomes and cost analysis, but this information is seldom captured in the structured electronic health record (EHR) data. The objective of this study was to develop and validate an artificial intelligence model to identify non-small cell lung cancer (NSCLC) patients from a cohort of heterogeneous LC patients using de-identified retrospective EHR data.

Read More

AI Model to Predict Cardiac AEs

Many oncology treatments have been associated with cardiovascular (CV) adverse events. We created a machine learning model to predict potential CV events in PD-(L)1 patients using the CancerLinQ database.

Read More

Risk Prediction Model

ECOG Imputation Modeling

AI Model to Impute ECOG Scores

COG PS is a prognostic indicator of outcomes, and scores of 0-1 (good ECOG PS) are often required for clinical trial enrollment. Patients treated in non-trial settings often lack ECOG PS scores limiting the ability of Real World Data from these patients to be used in external control arms (ECAs) or to provide optimal specificity for clinical effectiveness research. We developed a series of machine learning models to impute ECOG PS at initial diagnosis, metastatic diagnosis and final evaluation.

Read More

Predicting Survival at Specific Time Intervals

Survival prediction models for lung cancer patients could help guide their care and therapy decisions. We built a machine learning model to predict probability of survival beyond 90, 180 and 360 days from any point in a lung cancer patient’s journey.

Read More


 

Depiction of a patients journey as captured in the EHR and the features used to build the model

Depiction of a patient journey and features used to predict survival

AI Model to Improve Capture of Metastatic Status from EMR

Though an important prognostic feature in cancer, stage information is often missing from patient’s EHRs and unavailable in claims data. We developed and validated an AI model that classifies metastatic status in BC patients at their last observed timepoint (proxy for present-day).

Read More