If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Survival rates of lung cancer patients were rather poor until recent decades, when screening protocols, diagnostic techniques improvement and novel therapeutic options were developed. This leads to a new challenge: to increase lung cancer patients’ post-treatment quality of life (QoL) and well-being. We here report on a first integration of an NLP framework for the analysis and integration of comprehensive eElectronic Health Records, genomic data, open data sources, wearable devices and QoL questionnaires, in order to determine the factors that predict poor health status and design personalized interventions that will improve the patient's QoL.
Patients diagnosed and treated at the Medical Oncology Department at Puerta de Hierro University Hospital were included. Eligible patients were aged >18 years old, were diagnosed with non-small cell lung cancer (all stages), and had an ECOG 0-1. Artificial Intelligence (AI) and Knowledge Discovery (KD) techniques were used to integrate heterogeneous datasets, and synthesize complex relationships within these large data sets.
A total 2052 patients were included in the study. 251.730 documents from EHR were analyzed (240.851 notes and 10.879 reports) and images from patients have been included. A total of 124 patients wore the wearable device “Kronowise 3.0” (Kronohealth SL, Spain) and QoL questionnaires were also obtained from every patient. From every patient monitoring, more than 1.000.000 data records are being analyzed, and more than 130 indicators are obtained by using expert knowledge. These heterogeneous data sources are analyzed and integrated into an interactive user interface (Figure 1). This dashboard will allow clinicians to obtain immediate and personalized information of each patient and will elaborate models based on statistical relational learning and explainable AI techniques to predict patient-specific risk of developing complications and toxicities secondary to their cancer treatments. These models will help clinicians to make evidence-based treatment and post-treatment decisions in a way that it is not possible with any existing approach.
By using AI techniques we will be able to exploit large amounts of clinical information integrated into an interactive user that will facilitate the early discovery of risk factors that may deteriorate a lung cancer patient's condition during and after treatment. It will also allow us to examine the effect of multidisciplinary interventions in order to personalize their follow-up by better assessment of their needs and eventually improve their quality of life, wellbeing, and outcome. This work was supported by the EU H2020 program, under grant agreement Nº 875160 (Project CLARIFY).
Artificial Intelligence, personalized follow-up care, knowledge discovery