Granted Patent B2

US 12,582,353 · App. 17/899,194 · Granted Mar 24, 2026

Inventors: Elias Abou Zeid (Methuen, MA); Hillol Sarker (Belmont, MA); Archana Vaithilingam (Waban, MA); Dina Vogt (Belle Mead, NJ)

Assignee: Sanofi

A61B5/4848A61B5/0022A61B5/11A61B5/4818A61B5/6802A61B5/7203G16H10/60G16H20/10

Abstract

Claims (69)

1 . A computer-implemented method comprising:

receiving, by a computing system, one or more biomarker baselines for a patient, the biomarker baselines representing biological characteristics of the patient before going through a medical treatment;

receiving, by the computing system and from a client device, one or more signals representing respective biomarkers of the patient measured over a period of time subsequent to starting the medical treatment, wherein the signals are generated by one or more wearable sensors that are worn by the patient to measure the respective biomarkers over the period of time;

analyzing, by the computing system, the signals to obtain the respective biomarkers, wherein the analyzing comprises:

applying AI models on the one or more signals to obtain the respective biomarkers by estimating biomarkers based on patterns detected in the one or more signals over the period of time, wherein at least one of the AI models comprises an algorithm to determine (i) a left step or a right step of the patient and (ii) a step duration of the left step or the right step based on the one or more signals;

determining efficacy of the medical treatment by

comparing each of the obtained biomarkers with one or more respective biomarker baselines to determine respective changes in the biomarkers, and

comparing the respective changes with respective threshold values to determine the efficacy of the medical treatment on the patient; and

storing the determined efficacy for the medical treatment.

2 . The method of claim 1 , wherein the respective biomarkers comprise two or more of sleep apnea, gait, or physical activity.

3 . The method of claim 1 , wherein at least one signal includes data measured continuously over hours each day during the period of time.

4 . The method of claim 1 , wherein the medical treatment is a treatment for pediatric achondroplasia patients.

5 . The method of claim 1 , wherein the medical treatment comprises consuming a drug.

6 . The method of claim 5 , further comprising:

receiving respective measurement signals from multiple client devices respectively worn by multiple patients;

determining respective biomarkers for each of the multiple patients based on the respective measurement signals;

determining one or more dosages of the drug that causes the most changes in the respective biomarkers in at least some of the multiple patients;

determining, from among the one or more dosages, a particular dosage that causes the most changes in the respective biomarkers on most of patients in the multiple patients; and

storing or reporting the particular dosage as a recommended dosage for the drug.

7 . The method of claim 1 , wherein the respective biomarkers include sleep apnea,

wherein the method further comprises calculating an apnea hypopnea index based on the received signals, and

wherein determining the efficacy of the medical treatment further comprises;

determining a change in the apnea hypopnea index, and

marking the medical treatment as effective in response to determining a decrease in the apnea hypopnea index for more than a particular threshold value.

8 . The method of claim 1 , wherein the respective biomarkers include gait or physical activity assessments that are measured based on one or more of gait speed, step length, step duration, or physical activity measure of the patient.

9 . The method of claim 1 , wherein analyzing the signals further comprises:

filtering each signal in the received signals to cleanse the signal for a particular range of frequencies to generate a respective cleansed signal,

wherein the AI models are applied on the respective cleansed signal.

10 . The method of claim 1 , wherein the client device is remote from the computing system, and the signals are received from the client device through one or more wireless communication links.

11 . The method of claim 1 , wherein the client device includes at least one sensor in the one or more wearable sensors.

12 . The method of claim 1 , further comprising modulating the medical treatment based on the determined efficacy.

13 . A system comprising:

one or more computers; and

one or more computer-readable storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:

receiving, one or more biomarker baselines for a patient, the biomarker baselines representing biological characteristics of the patient before going through a medical treatment;

receiving, from a client device, one or more signals representing respective biomarkers of the patient measured over a period of time subsequent to starting the medical treatment, wherein the signals are generated by one or more wearable sensors that are worn by the patient to measure the respective biomarkers over the period of time;

analyzing the signals to obtain the respective biomarkers, wherein the analyzing comprises:

applying AI models on the one or more signals to obtain the respective biomarkers by estimating biomarkers based on patterns detected in the one or more signals over the period of time, wherein at least one of the AI models comprises an algorithm to determine (i) a left step or a right step of the patient and (ii) a step duration of the left step or the right step based on the one or more signals;

determining efficacy of the medical treatment by

comparing each of the obtained biomarkers with one or more respective biomarker baselines to determine respective changes in the biomarkers, and

comparing the respective changes with respective threshold values to determine the efficacy of the medical treatment on the patient; and

storing the determined efficacy for the medical treatment.

14 . The system of claim 13 , wherein the client device is remote from the system, and the signals are received from the client device through one or more wireless communication links.

15 . The system of claim 13 , wherein the respective biomarkers comprise two or more of sleep apnea, gait, or physical activity.

16 . The system of claim 13 , wherein the medical treatment comprises consuming a drug, and the operations further comprise;

receiving respective measurement signals from multiple client devices respectively worn by multiple patients;

determining respective biomarkers for each of the multiple patients based on the respective measurement signals;

determining one or more dosages of the drug that causes the most changes in the respective biomarkers in each at least some of the multiple patients;

determining, from among the one or more dosages, a particular dosage that causes the most changes in the respective biomarkers on most of patients in the multiple patients; and

storing or reporting the particular dosage as a recommended dosage for the drug.

17 . A non-transitory, computer-readable medium storing one or more instructions that are executable by one or more computers and that when executed by the one or more computers cause the one or more computers to perform operations comprising:

receiving, one or more biomarker baselines for a patient, the biomarker baselines representing biological characteristics of the patient before going through a medical treatment;

receiving, from a client device, one or more signals representing respective biomarkers of the patient measured over a period of time subsequent to starting the medical treatment, wherein the signals are generated by one or more wearable sensors that are worn by the patient to measure the respective biomarkers over the period of time;

analyzing the signals to obtain the respective biomarkers, wherein the analyzing comprises:

applying AI models on the one or more signals to obtain the respective biomarkers by estimating biomarkers based on patterns detected in the one or more signals over the period of time, wherein at least one of the AI models comprises an algorithm to determine (i) a left step or a right step of the patient and (ii) a step duration of the left step or the right step based on the one or more signals;

determining efficacy of the medical treatment by

comparing each of the obtained biomarkers with one or more respective biomarker baselines to determine respective changes in the biomarkers, and

comparing the respective changes with respective threshold values to determine the efficacy of the medical treatment on the patient; and

storing the determined efficacy for the medical treatment.

18 . The non-transitory, computer-readable medium of claim 17 , wherein the respective biomarkers comprise two or more of sleep apnea, gait, or physical activity.

19 . The non-transitory, computer-readable medium of claim 17 , wherein the medical treatment comprises consuming a drug, and the operations further comprise:

receiving respective measurement signals from multiple client devices respectively worn by multiple patients;

determining respective biomarkers for each of the multiple patients based on the respective measurement signals;

determining one or more dosages of the drug that causes the most changes in the respective biomarkers in each at least some of the multiple patients;

determining, from among the one or more dosages, a particular dosage that causes the most changes in the respective biomarkers on most of patients in the multiple patients; and

storing or reporting the particular dosage as a recommended dosage for the drug.

20 . The non-transitory, computer-readable medium of claim 17 , wherein analyzing the signals further comprises

filtering each signal in the received signals to cleanse the signal for a particular range of frequencies to generate respective cleansed signals,

wherein the AI models are applied on the respective cleansed signal.

Assignments (1)

Priority Claims (1)

Continuity (3)

References Cited (11)