Method and system for digital biomarkers platform

Patent metadata

Field	Value
Publication number	US-2022344055-A1
Application number	US-202217653248-A
Country	US
Kind code	A1
Filing date	Mar 2, 2022
Priority date	Mar 20, 2021
Publication date	Oct 27, 2022
Grant date	—

Abstract

Non-communicable diseases (NCDs) are the pandemics of modern era and are generating huge impact in the modern society. Conventional methods are inaccurate due to a challenge in handling data from heterogenous sensors. The present disclosure is capable of tracking fitness parameters of a user even with heterogenous sensors. Initially, the system receives a raw data from a plurality of heterogenous sensors associated with the user. The raw data is further transformed into a metadata format associated with the corresponding sensor. The transformed data is temporally aligned based on a time based slotting. An algorithm pipeline corresponding to a disorder to be analyzed is selected from a Directed Acyclic Graph (DAG) based on a sensor metadata and a plurality of algorithm metadata corresponding to a plurality of algorithms stored in an algorithm database and an algorithm pipeline. The corresponding disorder is analyzed using the algorithm pipeline.

First claim

What is claimed is:

1 . A processor implemented method comprising:
receiving, by one or more hardware processors, a raw data from a plurality of sensors associated with a user, wherein the plurality of sensors are heterogeneous; transforming, by the one or more hardware processors, the raw data corresponding to each of the plurality of sensors into a sensor metadata format associated with a corresponding sensor among the plurality of sensors; obtaining, by the one or more hardware processors, a temporally aligned data by performing a time based slotting on the transformed data; obtaining, by the one or more hardware processors, the sensor metadata corresponding to the temporally aligned data from a database based on a mapping between the temporally aligned data and the sensor metadata; selecting, by the one or more hardware processors, an algorithm pipeline corresponding to a disorder to be analyzed from a Directed Acyclic Graph (DAG), wherein the selection is based on the sensor metadata and a plurality of algorithm metadata corresponding to a plurality of algorithms stored in an algorithm database; and analyzing, by the one or more hardware processors, the disorder associated with the user based on the selected algorithm pipeline.

2 . The method of claim 1 , wherein selecting the algorithm pipeline corresponding to the selected disorder to be analyzed from the DAG based on the sensor metadata and a plurality of algorithm metadata comprises:
receiving the DAG comprising a plurality of nodes and a plurality of edges, wherein each of the plurality of nodes associated with the DAG represents an algorithm from the plurality of algorithms and each of the plurality of edges connecting two nodes indicates a dependency between a preceding node and a succeeding node, wherein the DAG is associated with at least one entry point; selecting an initial node corresponding to a first algorithm of the algorithm pipeline from the DAG based on a comparison between the corresponding sensor metadata and an algorithm metadata associated with each of the plurality of algorithms, wherein the initial node is assigned as a current node; obtaining the algorithm metadata associated with a plurality of next nodes associated with the current node; selecting a next node from the plurality of next nodes based on a comparison between an output associated with the current node and the algorithm schema associated with each of the plurality of next nodes, wherein the selected next node is assigned as the current node, and wherein the DAG is traversed until reaching a leaf node identified based on a next null pointer; and obtaining the algorithm pipeline based on the traversal path from the initial node to the leaf node.

3 . The method of claim 1 , wherein constructing the DAG for a disorder from a plurality of disorders comprises:
Receiving a first algorithm corresponding to the disorder, the algorithm metadata associated with the first algorithm, an output associated with the first algorithm and the corresponding sensor metadata, wherein the output associated with the first algorithm is predetermined; Inserting an initial node into the DAG corresponding to the first algorithm based on a mapping between the algorithm metadata associated with the first algorithm and the corresponding sensor metadata associated with the corresponding disorder, wherein the first algorithm is marked as visited, and wherein the initial node is assigned as a current node; selecting a second algorithm from the plurality of algorithms based on a mapping between the output associated with the first algorithm, the corresponding sensor metadata and the algorithm metadata associated with a plurality of unvisited algorithms; generating a second node corresponding to the second algorithm when there is a mapping existing between the output associated with the first algorithm, the corresponding sensor metadata and at least one of the plurality of unvisited algorithms; and generating an edge between the current node and the second node, wherein the second node is assigned as the current node, wherein constructing the DAG is performed until a plurality of measurements associated with the corresponding disorder is satisfied, wherein the plurality of measurements comprises a heart-rate of the user, a step count and a calorie.

4 . The method of claim 1 , wherein the raw data format comprises a time series data, a comma separated values, a string data, a numerical, a Boolean data, a list, a plurality of objects, a plurality of videos and a plurality of images.

5 . The method of claim 1 , wherein the algorithm metadata comprising a data format, a preferred sampling rate, a minimum quantum of data needed for processing, an output data format, a description of method used by the corresponding algorithm.

6 . The method of claim 1 , wherein the sensor metadata comprises a type of data generated by the corresponding sensor, a sampling frequency, a buffering capacity and a model associated with the corresponding sensor.

7 . A system comprising:
at least one memory storing programmed instructions; one or more Input/Output (I/O) interfaces; and one or more hardware processors operatively coupled to the at least one memory, wherein the one or more hardware processors are configured by the programmed instructions to: receive a raw data from a plurality of sensors associated with a user, wherein the plurality of sensors are heterogeneous; transform the raw data corresponding to each of the plurality of sensors into a sensor metadata format associated with a corresponding sensor among the plurality of sensors; obtain a temporally aligned data by performing a time based slotting on the transformed data; obtain the sensor metadata corresponding to the temporally aligned data from a database based on a mapping between the temporally aligned data and the sensor metadata; select an algorithm pipeline corresponding to a disorder to be analyzed from a Directed Acyclic Graph (DAG), wherein the selection is based on the sensor metadata and a plurality of algorithm metadata corresponding to a plurality of algorithms stored in an algorithm database; and analyze the disorder associated with the user based on the selected algorithm pipeline.

8 . The system of claim 7 , wherein selecting the algorithm pipeline corresponding to the selected disorder to be analyzed from the DAG based on the sensor metadata and a plurality of algorithm metadata comprises:
receiving the DAG comprising a plurality of nodes and a plurality of edges, wherein each of the plurality of nodes associated with the DAG represents an algorithm from the plurality of algorithms and each of the plurality of edges connecting two nodes indicates a dependency between a preceding node and a succeeding node, wherein the DAG is associated with at least one entry point; selecting an initial node corresponding to a first algorithm of the algorithm pipeline from the DAG based on a comparison between the corresponding sensor metadata and an algorithm metadata associated with each of the plurality of algorithms, wherein the initial node is assigned as a current node; obtaining the algorithm metadata associated with a plurality of next nodes associated with the current node; selecting a next node from the plurality of next nodes based on a comparison between an output associated with the current node and the algorithm schema associated with each of the plurality of next nodes, wherein the selected next node is assigned as the current node, and wherein the DAG is traversed until reaching a leaf node identified based on a next null pointer; and obtaining the algorithm pipeline based on the traversal path from the initial node to the leaf node.

Assignees

• Tata Consultancy Services Ltd

Inventors

• Ghose Avik
• Samal Avijit
• Ahmed Nasimuddin
• Singhal Shivam
• Bhavsar Karan
• Chandel Vivek
• Khandelwal Sundeep
• Vishwakarma Harsh
• Pawar Bhaskar

Classifications

• G16H20/30

  relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising · CPC title

• G16H50/30 Primary

  for calculating health indices; for individual health risk assessment · CPC title

• G16H50/20

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• G16H40/63 Primary

  for local operation · CPC title

• G16H40/67

  for remote operation · CPC title