Method and system enabling digital biomarker data integration and analysis for clinical treatment impact
The instant system and method will be a scalable and flexible software platform to agnostically capture wearable, implantable, or external device data and create a digital biomarker for various disease conditions, care and analysis. The system and method also provides an informatics tool for automated data aggregation, integration, and machine learning algorithms to uniformly compare/contrast/couple data streams as a function of patient and context over time. The focus is on user engagement and usability, accessibility and reporting to the clinical care team, and usability by researchers interested in analyzing data and outcomes.
1. A method for integration and analysis for a clinical treatment impact, the method comprising:
receiving, by a platform, digital biomarker data from a wearable device of a patient;
forming, by the platform and utilizing a clustered aGgregation (CAG) algorithm, a cluster of mobile devices that includes a mobile device associated with the patient, wherein the forming is based on a determination that sensor data, for each mobile device in the cluster, is within a given threshold range over a predetermined time;
receiving, by the platform, a subset of the sensor data from the cluster of mobile devices;
providing a plurality of different types of digital data, obtained at least in part from the mobile device associated with the patient at a particular point in time, as input to one or more neural network models, wherein particular digital data, of the plurality of different types of digital data, is an integer value that represents a category of a plurality of different categories and the particular digital data is transformed into a binary vector before being provided as input to the one or more neural network models,
wherein the plurality of different types of digital data include at least the subset of sensor data, voice data from the mobile device, and camera data from the mobile device;
determining, by a prediction unit, a physical activity being performed by the patient at the particular point in time and a situational context for the physical activity being performed by the patient at the particular point in time, wherein
the determining is based on an output of the one or more neural network models,
the physical activity is one of a plurality of different types of physical activities, and
the situational context indicates a type of physical environment, of a plurality of different types of physical environments, in which the physical activity is being performed;
integrating, by the platform, the digital biomarker data with one or more of the plurality of different types of digital data to generate combined data, wherein the integrating generates a machine learning model and analysis projection using a cloud based analysis engine to integrate the digital biomarker data with the one or more of the plurality of different types of digital data to generate the combined data; and
generating, by the platform, one or more clinical condition predictions for the patient by utilizing the combined data, the determined activity, and the situational context for the physical activity with one or more prediction algorithms.
2. A system for a digital data integration and analysis for a clinical treatment impact, comprising:
a platform configured to:
receive digital biomarker data from wearable device of a patient;
form, utilizing a clustered aGgregation (CAG) algorithm, a cluster of mobile devices that includes a mobile device associated with the patient, wherein the forming is based on a determination that sensor data, for each mobile device in the cluster, is within a given threshold range over a predetermined time;
receive a subset of the sensor data from the cluster of mobile devices;
provide a plurality of different types of digital data, obtained at least in part from the mobile device associated with the patient at a particular point in time, as input to one or more neural network models, wherein particular digital data, of the plurality of different types of digital data, is an integer value that represents a category of a plurality of different categories and the particular digital data is transformed into a binary vector before being provided as input to the one or more neural network models,
wherein the plurality of different types of digital data include at least the subset of sensor data, voice data from the mobile device, and camera data from the mobile device;
determine, based on an output of the one or more neural network models, a physical activity being performed by the patient at the particular point in time and a situational context for the physical activity being performed by the patient at the particular point in time, wherein
the physical activity is one of a plurality of different types of physical activities; integrate the digital biomarker data with one or more of the plurality of different types of digital data to generate combined data, and
the situational context indicates a type of physical environment, of a plurality of different types of physical environments, in which the physical activity is being performed; and
generate one or more clinical condition predictions for the patient by utilizing the combined data, the determined physical activity, and the situational context with one or more prediction algorithms.
3. The method of claim 1 ,
wherein the machine learning model and analysis projection further utilizes one or more of (1) supervised and unsupervised modelling, (2) dimensionality reduction, (3) discriminative modelling, or (4) and generative modelling.
4. The method of claim 1 , further comprising:
integrating the digital biomarker data from the wearable device with different digital biomarker data from other wearable devices associated with the patient to generate combined digital biomarker data.
5. The method of claim 1 , further comprising:
utilizing the digital biomarker data in conjunction with existing medical data associated with the patient for management of one or more of a symptom, side effect, a quality of life indicator, or a disease progression for the patient.
6. The method of claim 1 , further comprising:
determining (1) one or more changes in a performance status of the patient and (2) one or more changes in a drug treatment provided to the patient throughout a clinical trial.
7. The method of claim 1 , wherein the sensor data comprises one or more of accelerometer data or gyroscope data.
8. The method of claim 1 , further comprising:
receiving, by the platform, digital medical data associated with the patient; and
integrating the digital biomarker data, the digital data, and the digital medical data to generate the combined data.
9. The method of claim 8 , wherein the digital medical data is provided via one or more user interfaces of an application executing on the mobile device associated with the patient.
10. A system for a digital data integration and analysis for a clinical treatment impact, comprising:
a platform configured to:
receive digital biomarker data from wearable device of a patient;
form, utilizing a clustered aGgregation (CAG) algorithm, a cluster of mobile devices that includes a mobile device associated with the patient, wherein the forming is based on a determination that sensor data, for each mobile device in the cluster, is within a given threshold range over a predetermined time;
receive a subset of the sensor data from the cluster of mobile devices;
provide a plurality of different types of digital data, obtained from the mobile device at a particular point in time, as input to one or more neural network models,
wherein the plurality of different types of digital data include at least the subset of sensor data, voice data from the mobile device, and camera data from the mobile device;
determine, based on an output of the one or more neural network models, a physical activity being performed by the patient at the particular point in time and a situational context for the physical activity being performed by the patient at the particular point in time, wherein
the physical activity is one of a plurality of different types of physical activities; integrate the digital biomarker data with one or more of the plurality of different types of digital data to generate combined data, and
the situational context indicates a type of physical environment, of a plurality of different types of physical environments, in which the physical activity is being performed; and
generate one or more clinical condition predictions for the patient by utilizing the combined data, the determined physical activity, and the situational context with one or more prediction algorithms.
11. The system of claim 10 , wherein the platform is further configured to obtain digital medical data associated with the patient.
12. The system of claim 11 , wherein the platform is further configured to integrate the digital biomarker data, the digital data, and the medical data to generate the combined data.
13. The system of claim 11 , wherein the digital medical data is provided via one or more user interfaces of an application executing on the mobile device associated with the patient.
14. The system of claim 10 , wherein the platform is further configured to:
generate a machine learning model and analysis projection using a cloud based analysis engine, wherein the machine learning model and analysis projection utilizes one or more of (1) supervised and unsupervised modelling, (2) dimensionality reduction, (3) discriminative modelling, or (4) and generative modelling.
15. The system of claim 10 , further comprising:
utilize the digital biomarker data in conjunction with existing medical data for management of one or more of a symptom, a side effect, a quality of life indicator, or a disease progression.
16. The system of claim 10 , wherein the sensor data comprises one or more of accelerometer data or gyroscope data.