• Development and validation of...

Treffer: Development and validation of deep learning for predicting the growth of ovarian cancer organoids.

Title:
Development and validation of deep learning for predicting the growth of ovarian cancer organoids.
Authors:
Wu H; Bioengineering College, Chongqing University, Chongqing 400044, China., Ma L; Department of Gynecologic Oncology, Chongqing University Cancer Hospital & Chongqing Cancer Institute & Chongqing Cancer Hospital, Chongqing 400030, China.; Chongqing Specialized Medical Research Center of Ovarian Cancer, Chongqing 400030, China.; Organoid Transformational Research Center, Chongqing Key Laboratory of Translational Research for Cancer Metastasis and Individualized Treatment, Chongqing University Cancer Hospital, Chongqing 400030, China., Wang L; Department of Gynecologic Oncology, Chongqing University Cancer Hospital & Chongqing Cancer Institute & Chongqing Cancer Hospital, Chongqing 400030, China.; Chongqing Specialized Medical Research Center of Ovarian Cancer, Chongqing 400030, China.; Organoid Transformational Research Center, Chongqing Key Laboratory of Translational Research for Cancer Metastasis and Individualized Treatment, Chongqing University Cancer Hospital, Chongqing 400030, China., Zhu X; Department of Gynecologic Oncology, Chongqing University Cancer Hospital & Chongqing Cancer Institute & Chongqing Cancer Hospital, Chongqing 400030, China.; Chongqing Specialized Medical Research Center of Ovarian Cancer, Chongqing 400030, China.; Organoid Transformational Research Center, Chongqing Key Laboratory of Translational Research for Cancer Metastasis and Individualized Treatment, Chongqing University Cancer Hospital, Chongqing 400030, China., Luo X; Bioengineering College, Chongqing University, Chongqing 400044, China., Zhang C; School of Medicine, Chongqing University, Chongqing 400044, China., Ha C; Gynecology Department, General Hospital of Ningxia Medical University, Yinchuan, Ningxia 750003, China., Dang Y; Gansu Provincial Maternity and Child Care Hospital, Lanzhou, Gansu 730050, China.; Institute of Pathology, School of Basic Medical Sciences, Lanzhou University, Lanzhou, Gansu 730000, China., Wang H; Department of Gynecologic Oncology, Chongqing University Cancer Hospital & Chongqing Cancer Institute & Chongqing Cancer Hospital, Chongqing 400030, China.; Chongqing Specialized Medical Research Center of Ovarian Cancer, Chongqing 400030, China.; Organoid Transformational Research Center, Chongqing Key Laboratory of Translational Research for Cancer Metastasis and Individualized Treatment, Chongqing University Cancer Hospital, Chongqing 400030, China., Zou D; Department of Gynecologic Oncology, Chongqing University Cancer Hospital & Chongqing Cancer Institute & Chongqing Cancer Hospital, Chongqing 400030, China.; Chongqing Specialized Medical Research Center of Ovarian Cancer, Chongqing 400030, China.; Organoid Transformational Research Center, Chongqing Key Laboratory of Translational Research for Cancer Metastasis and Individualized Treatment, Chongqing University Cancer Hospital, Chongqing 400030, China.
Source:
Chinese medical journal [Chin Med J (Engl)] 2026 Jan 05; Vol. 139 (1), pp. 108-117. Date of Electronic Publication: 2025 Jul 25.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: Chinese Medical Association ; produced by Wolters Kluwer Country of Publication: China NLM ID: 7513795 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 2542-5641 (Electronic) Linking ISSN: 03666999 NLM ISO Abbreviation: Chin Med J (Engl) Subsets: MEDLINE
Imprint Name(s):
Publication: <2015- > : Beijing : Chinese Medical Association ; produced by Wolters Kluwer
Original Publication: Peking, Chinese Medical Assn.
MeSH Terms:
Ovarian Neoplasms*/diagnosis , Ovarian Neoplasms*/pathology , Microscopy*/methods , Predictive Learning Models* , Deep Learning* , Organoids*/growth & development, Humans ; Female ; Software Validation ; Longitudinal Studies ; Area Under Curve ; Reproducibility of Results ; Predictive Value of Tests
References:
Nat Commun. 2021 May 10;12(1):2614. (PMID: 33972525)
Comput Methods Programs Biomed. 2020 Jul;190:105351. (PMID: 32028084)
Cell Rep. 2018 Jul 31;24(5):1363-1376. (PMID: 30067989)
Nat Rev Cancer. 2018 Jul;18(7):407-418. (PMID: 29692415)
Nat Mach Intell. 2023;5(8):933-946. (PMID: 37615030)
J Digit Imaging. 2023 Apr;36(2):441-449. (PMID: 36474087)
BMC Med Imaging. 2022 Apr 13;22(1):69. (PMID: 35418051)
Sci Rep. 2019 Aug 28;9(1):12479. (PMID: 31462669)
J Biomed Inform. 2023 Aug;144:104458. (PMID: 37488023)
Nat Commun. 2021 Sep 28;12(1):5678. (PMID: 34584080)
Cell Oncol (Dordr). 2022 Oct;45(5):1019-1036. (PMID: 36036881)
Cell Syst. 2023 Sep 20;14(9):764-776.e6. (PMID: 37734323)
Lab Chip. 2022 Oct 25;22(21):4118-4128. (PMID: 36200406)
Cell. 2014 Sep 25;159(1):176-187. (PMID: 25201530)
JAMA. 2016 Dec 13;316(22):2402-2410. (PMID: 27898976)
Cell Stem Cell. 2023 Oct 5;30(10):1351-1367.e10. (PMID: 37802039)
Curr Biol. 2017 Sep 25;27(18):R1024-R1035. (PMID: 28950087)
Tissue Eng Regen Med. 2023 Dec;20(7):1109-1117. (PMID: 37594633)
Comput Biol Med. 2021 Jul;134:104490. (PMID: 34102401)
Comput Methods Programs Biomed. 2022 Nov;226:107161. (PMID: 36228495)
Nature. 2015 May 28;521(7553):436-44. (PMID: 26017442)
IEEE J Biomed Health Inform. 2023 Jul;27(7):3360-3371. (PMID: 37099473)
Dev Cell. 2023 Jun 19;58(12):1106-1121.e7. (PMID: 37148882)
Comput Methods Programs Biomed. 2023 Feb;229:107264. (PMID: 36473419)
Front Cell Neurosci. 2020 Jul 03;14:171. (PMID: 32719585)
Cancer Sci. 2022 Aug;113(8):2693-2703. (PMID: 35585758)
Front Bioeng Biotechnol. 2022 Jan 27;9:802794. (PMID: 35155409)
Cell. 2015 Jan 15;160(1-2):324-38. (PMID: 25557080)
Commun Biol. 2021 Mar 26;4(1):415. (PMID: 33772211)
Neuroimage. 2017 Feb 1;146:1038-1049. (PMID: 27693612)
Nat Commun. 2024 Mar 27;15(1):2681. (PMID: 38538600)
Med. 2021 Sep 10;2(9):1011-1026. (PMID: 34617071)
Sci Rep. 2018 Jan 29;8(1):1727. (PMID: 29379060)
Cell. 2015 May 7;161(4):933-45. (PMID: 25957691)
Math Biosci Eng. 2020 Sep 15;17(5):6203-6216. (PMID: 33120595)
Cancer Discov. 2018 Nov;8(11):1404-1421. (PMID: 30213835)
Stem Cell Reports. 2019 Apr 9;12(4):845-859. (PMID: 30880077)
IEEE Trans Image Process. 2021;30:7025-7037. (PMID: 34329165)
Comput Methods Programs Biomed. 2023 May;233:107466. (PMID: 36907040)
Med Image Anal. 2017 Dec;42:60-88. (PMID: 28778026)
Cell Rep. 2021 Jul 27;36(4):109429. (PMID: 34320344)
Comput Biol Med. 2023 Sep;164:107313. (PMID: 37562325)
Front Bioeng Biotechnol. 2023 Apr 12;11:1133090. (PMID: 37122853)
Contributed Indexing:
Keywords: Artificial intelligence; Deep learning; Organoid; Ovarian cancer; Precision medicine
Entry Date(s):
Date Created: 20250725 Date Completed: 20260105 Latest Revision: 20260107
Update Code:
20260107
PubMed Central ID:
PMC12768030
DOI:
10.1097/CM9.0000000000003575
PMID:
40709801
Database:
MEDLINE

Weitere Informationen

Background: Organoids have attracted enormous interest in disease modeling, drug screening, and precision medicine. However, developing robust patient-derived organoids (PDOs) was time-consuming, costly, and had low success rates for certain cancer types, which limited their clinical utility. This study aimed to develop an interpretable deep learning-based model to predict the cultivation outcome of ovarian cancer organoids in advance.
Methods: Longitudinal microscopy images of 517 ovarian cancer organoid droplets were divided into training ( n = 325), validation ( n = 88), and test ( n = 104) sets. Subsequently, growth prediction models were developed based on four neural network backbones (ResNet18, VGG11, ConvNeXt v2, and Swin Transformer v2), and specific optimization methods were designed for better prediction. Finally, 179 samples from multiple centers were collected for prospective validation, and the gradient-weighted class activation mapping (Grad-CAM) method was used for interpretability analysis of the deep model to reveal the basis of the model's decisions.
Results: The test set showed that the deep learning models could achieve high-performance prediction at the third stage with area under the curve (AUC) values greater than 0.8 for all four models. The homogeneous transfer learning optimization method improved the AUC from 0.833 to 0.884 ( P = 0.0039). In prospective validation, the optimized model achieved an AUC of 0.832, a Brier score of 0.1919 in the calibration curve, and a greater net benefit in the decision curve. Interpretability analysis revealed that the area where organoids are being formed and have already formed is important for prediction.
Conclusions: Our developed models achieved satisfactory results in predicting the growth of ovarian cancer organoids. There is potential for further development of the model toward process automation.
(Copyright © 2025 The Chinese Medical Association, produced by Wolters Kluwer, Inc. under the CC-BY-NC-ND license.)