IP Library › Granted Patent US 12,456,213
Granted Patent B2
US 12,456,213 · App. 18/203,922 · Granted Oct 28, 2025

Generating views of skin lesions for improved remote evaluation

Inventors: Abhivyakti Sawarkar (Cambridge, MA); Ayushi Sinha (Cambridge, MA)
Assignee: Koninklijke Philips N.V.
G06T7/55G06T2207/20081G06T2207/30096
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,456,213
App. No.
18/203,922
Granted
Oct 28, 2025
Kind
B2
Abstract

A method for generating a continuous digital volumetric representation of a region of a patient, comprising: (i) receiving, by a volumetric representation generation system, an instruction or request to generate a digital volumetric representation of the region of the patient; (ii) providing one or more guidance instructions regarding obtaining a plurality of images of the region; (iii) receiving the plurality of images, wherein at least of each of two or more of the plurality of images are obtained of the region of the patient from a different angle; (iv) processing, using a trained model, the received plurality of images to generate a continuous digital volumetric representation of the region; and (v) providing the generated continuous digital volumetric representation of the region to a user.

Claims (36)

1. A method for generating a continuous digital volumetric representation of a region of a patient, comprising:

receiving by a volumetric representation generation system, an instruction or request to generate a digital volumetric representation of the region of the patient;

providing, to the patient via a user interface, one or more guidance instructions regarding obtaining a plurality of images of the region of the patient;

receiving a plurality of images obtained of the region of the patient, wherein at least of each of two or more of the plurality of images are obtained of the region of the patient from a different angle;

processing using a trained model, the received plurality of images to generate a continuous digital volumetric representation of the region;

providing the generated continuous digital volumetric representation of the region to a user;

processing one or more generated continuous digital volumetric representations of the region, the one or more generated continuous digital volumetric representations of the region generated from a plurality of images of the region of the patient obtained at one or more different times; and

computing, based on the one or more generated continuous digital volumetric representations of the region, one or more clinically-relevant metrics and/or risks.

2. The method of claim 1 , wherein the trained model is further configured to generate, by processing the received plurality of images, a confidence level for the generated continuous digital volumetric representation of the region.

3. The method of claim 1 , further comprising the steps of:

providing, to the patient via the user interface, an instruction to obtain one or more images of the region of the patient, wherein the instruction is based on the generated continuous digital volumetric representation of the region;

receiving one or more images of the region of the patient; and re-processing, using the trained model, the received plurality of images and the received one or more images to generate an updated continuous digital volumetric representation of the region.

4. The method of claim 1 , wherein the plurality of images obtained of the region of the patient are obtained using a digital camera.

5. The method of claim 4 , wherein the digital camera is a component of a wearable device or a smartphone of the patient.

6. The method of claim 1 , further comprising training the model to process a received plurality of images of a region to generate at least a continuous digital volumetric representation of the region.

7. A method, comprising:

receiving a generated continuous digital volumetric representation of a region of a patient, wherein the continuous digital volumetric representation of the region of the patient is generated by the process:

receiving, by the system, an instruction or request to generate a digital volumetric representation of the region of the patient;

providing to the patient via a user interface, one or more guidance instructions regarding obtaining a plurality of images of the region of the patient;

receiving a plurality of images obtained of the region of the patient, wherein at least of each of two or more of the plurality of images are obtained of the region of the patient from a different angle;

processing, using a trained model, the received plurality of images to generate the continuous digital volumetric representation of the region; and

manipulating, using a visualization controller and user interface, the received continuous digital volumetric representation, wherein manipulation comprises at least viewing the received continuous digital volumetric representation from two or more different angles; and

processing one or more generated continuous digital volumetric representations of the region; and

computing, based on the one or more generated continuous digital volumetric representations of the region, one or more clinically-relevant metrics and/or risks.

8. The method of claim 7 , wherein the plurality of images obtained of the region of the patient are obtained using a digital camera.

9. The method of claim 8 , wherein the digital camera is a component of a wearable device or a smartphone of the patient.

10. The system of claim 9 , wherein the processor is further configured to train the volumetric representation model to process a received plurality of images of a region to generate at least a continuous digital volumetric representation of the region.

11. A system for generating a continuous digital volumetric representation of a region of a patient, comprising:

a plurality of images obtained of a region of a patient, wherein at least of each of two or more of the plurality of images are obtained of the region of the patient from a different angle;

a trained volumetric representation model;

a processor configured to analyze, using the trained volumetric representation model, the received plurality of images to generate a continuous digital volumetric representation of the region;

a user interface configured to provide the generated continuous digital volumetric representation of the region to a user, wherein the processor is further configured to process the one or more generated continuous digital volumetric representations of the region, the one or more generated continuous digital volumetric representations of the region generated from a plurality of images of the region of the patient obtained at one or more different times, and compute, based on the one or more generated continuous digital volumetric representations of the region, one or more clinically-relevant metrics and/or risks.

12. The system of claim 11 , wherein the processor is further configured to receive an instruction or request to generate a digital volumetric representation of the region of the patient, and wherein the user interface is further configured to provide one or more guidance instructions regarding obtaining a plurality of images of the region of the patient.

13. The system of claim 11 , wherein the processor is further configured to: (i) provide an instruction to obtain one or more images of the region of the patient, wherein the instruction is based on the generated continuous digital volumetric representation of the region; (ii) receive one or more images of the region of the patient; and (iii) re-process, using the trained volumetric representation model, the received plurality of images and the received one or more images to generate an updated continuous digital volumetric representation of the region.

14. The system of claim 11 , wherein the plurality of images obtained of the region of the patient are obtained using a digital camera.

15. The system of claim 14 , wherein the digital camera is a component of a wearable device or a smartphone of the patient.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 31, 2023
From: SINHA, AYUSHI; SAWARKAR, ABHIVYAKTI
To: KONINKLIJKE PHILIPS N.V.
Reel/Frame 063810/0077 →
Continuity (2)
Provisional Application 63351571 · Jun 13, 2022
Related Publication 20230410339A1 · Dec 21, 2023
References Cited (23)
US 20140313303A1 · Davis · 2014 [cited by applicant]
US 20200226816A1 · Kar · 2020 [cited by applicant]
US 20200234444A1 · Budman · 2020 [cited by applicant]
US 20220005191A1 · Connor · 2022 [cited by applicant]
US 20220036050A1 · Chandler · 2022 [cited by applicant]
US 20220036602A1 · Duckworth · 2022 [cited by applicant]
US 20220047165A1 · Perkins · 2022 [cited by applicant]
US 20220157450A1 · Kozloski · 2022 [cited by applicant]
US 20220351465A1 · Pantpratinidhi · 2022 [cited by examiner]
CN 113706714A1 · 2021 [cited by applicant]
Jo, K. et al., “CG-NeRF: Conditional Generative Neural Radiance Fields”. Association for the Advancement of Artificial Intelligence (2022). [cited by applicant]
Lin, C-H et al., “BARF: Bundle-Adjusting Neural Radiance Fields”. arXiv:2104.06405 [cs.CV]. [cited by applicant]
Mildenhall, B. et al., “NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis”. arXiv:2003.08934 [cs.CV]. [cited by applicant]
Yu, A. et al., “pixelNeRF: Neural Radiance Fields from One or Few Images.” Proc. CVPR. May 30, 2021 [cited Sep. 16, 2021]; Available from: http://arxiv.org/abs/2012.02190. [cited by applicant]
Chuchu, N. et al., “Smartphone applications for triaging adults with skin lesions that are suspicious for melanoma”. Cochrane Database Syst Rev. Dec. 4, 2018:12:CD013192. PMCID: PMC6517294. [cited by applicant]
Chao, E. et al., “Smartphone-Based Applications for Skin Monitoring and Melanoma Detection.” Dermatol Clin. Oct. 2017;35(4):551-557. PMID: 28886812. [cited by applicant]
Phillips, M. et al., “Assessment of Accuracy of an Artificial Intelligence Algorithm to Detect Melanoma in Images of Skin Lesions.” JAMA Netw Open. Oct. 2, 2019;2(10):e1913436. PMCID: PMC6806667. [cited by applicant]
Ngoo, AI. et al., “Efficacy of smartphone applications in high-risk pigmented lesions.” Australas J Dermatol. Aug. 2018:59(3):e175-e182. PMID: 28240347. [cited by applicant]
Mar, V.J. et al., “Artificial intelligence for melanoma diagnosis: how can we deliver on the promise?” Ann Oncol. Aug. 1, 2018;29(8):1625-1628. PMID: 29846499. [cited by applicant]
Morley, J. et al., “The poor performance of apps assessing skin cancer risk.” BMJ. Feb. 10, 2020;368:m428. PMID: 32048610. [cited by applicant]
Rat, C. et al., “Use of Smartphones for Early Detection of Melanoma: Systematic Review.” J Med Internet Res. Apr. 13, 2018:20(4):e135. PMCID: PMC5923035. [cited by applicant]
International Search Report for PCT/EP2023/064853 filed Jun. 2, 2023. [cited by applicant]
Sun Shao-Hua et al., “Multi-view to Novel View: Synthesizing Novel Views with Self-learned Confidence”, (Ocy. 7, 2018), European Conference on Computer Vision, p. 162-178. [cited by applicant]