EHRs help to identify symptom clusters in cancer patients
Researchers from the Regenstrief Institute and Indiana University–Purdue University Indianapolis have developed novel methods for extracting data on patient symptoms from electronic health records.
Researchers from the Regenstrief Institute and Indiana University–Purdue University Indianapolis have developed novel methods for extracting data on patient symptoms from electronic health records.
By analyzing free text clinical notes and extracting information from structured entries in EHRs, investigators were able to identify symptom clusters—those symptoms that tend to go together and are associated with disease.
“EHR data has not been extensively used to understand patient-reported symptoms for individuals with chronic diseases,” says Xiao Luo, assistant professor of computer and information technology in IUPUI’s School of Engineering and Technology. “Utilizing EHR data obtained from the Indiana Network for Patient Care, we developed a framework that employs components of data mining, NLP and machine learning to explore clinical information accumulated throughout the course of these diseases.”
In a study, symptom clusters during two time periods—the first year of chemotherapy and the 48 to 54 months following chemotherapy—were identified in breast cancer and colorectal cancer patients.
What researchers found were that symptom clusters had different associations between breast cancer and colorectal cancer, as well as for different time frames after chemotherapy. For example, breast cancer patients had slightly more symptoms than colorectal cancer patients during the first year after chemotherapy.
“The colorectal cancer patient cohort has slightly more depression on average between 48 months and 54 months after the chemotherapy,” states the study, which won best paper at the 10th Association for Computing Machinery Conference on Bioinformatics, Computational Biology and Health Informatics.
“Through applying the association rule mining, we find some informative rules, such as ‘if a patient is at a higher cancer stage of colorectal cancer (3B), but no fatigue symptom, he or she likely doesn’t have depression and peripheral neuropathy’,” add the authors.
According to senior author and Regenstrief Institute investigator Kun Huang, the novel methods for extracting data on patient symptoms from EHRs “can be generalized beyond breast and colorectal cancer to analyze symptom clusters of other chronic diseases where symptom management and treatment is critical.”
By analyzing free text clinical notes and extracting information from structured entries in EHRs, investigators were able to identify symptom clusters—those symptoms that tend to go together and are associated with disease.
“EHR data has not been extensively used to understand patient-reported symptoms for individuals with chronic diseases,” says Xiao Luo, assistant professor of computer and information technology in IUPUI’s School of Engineering and Technology. “Utilizing EHR data obtained from the Indiana Network for Patient Care, we developed a framework that employs components of data mining, NLP and machine learning to explore clinical information accumulated throughout the course of these diseases.”
In a study, symptom clusters during two time periods—the first year of chemotherapy and the 48 to 54 months following chemotherapy—were identified in breast cancer and colorectal cancer patients.
What researchers found were that symptom clusters had different associations between breast cancer and colorectal cancer, as well as for different time frames after chemotherapy. For example, breast cancer patients had slightly more symptoms than colorectal cancer patients during the first year after chemotherapy.
“The colorectal cancer patient cohort has slightly more depression on average between 48 months and 54 months after the chemotherapy,” states the study, which won best paper at the 10th Association for Computing Machinery Conference on Bioinformatics, Computational Biology and Health Informatics.
“Through applying the association rule mining, we find some informative rules, such as ‘if a patient is at a higher cancer stage of colorectal cancer (3B), but no fatigue symptom, he or she likely doesn’t have depression and peripheral neuropathy’,” add the authors.
According to senior author and Regenstrief Institute investigator Kun Huang, the novel methods for extracting data on patient symptoms from EHRs “can be generalized beyond breast and colorectal cancer to analyze symptom clusters of other chronic diseases where symptom management and treatment is critical.”
More for you
Loading data for hdm_tax_topic #better-outcomes...