Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 May 29;7(1):143.
doi: 10.1038/s41746-024-01131-7.

Deep learning to assess microsatellite instability directly from histopathological whole slide images in endometrial cancer

Ching-Wei Wang  1 , Hikam Muzakky  1 , Nabila Puspita Firdi  1 , Tzu-Chien Liu  1 , Po-Jen Lai  1 , Yu-Chi Wang  2   3 , Mu-Hsien Yu  2   3 , Tai-Kuang Chao  4   5
Affiliations

Deep learning to assess microsatellite instability directly from histopathological whole slide images in endometrial cancer

Ching-Wei Wang et al. NPJ Digit Med. .

Abstract

Molecular classification, particularly microsatellite instability-high (MSI-H), has gained attention for immunotherapy in endometrial cancer (EC). MSI-H is associated with DNA mismatch repair defects and is a crucial treatment predictor. The NCCN guidelines recommend pembrolizumab and nivolumab for advanced or recurrent MSI-H/mismatch repair deficient (dMMR) EC. However, evaluating MSI in all cases is impractical due to time and cost constraints. To overcome this challenge, we present an effective and efficient deep learning-based model designed to accurately and rapidly assess MSI status of EC using H&E-stained whole slide images. Our framework was evaluated on a comprehensive dataset of gigapixel histopathology images of 529 patients from the Cancer Genome Atlas (TCGA). The experimental results have shown that the proposed method achieved excellent performances in assessing MSI status, obtaining remarkably high results with 96%, 94%, 93% and 100% for endometrioid carcinoma G1G2, respectively, and 87%, 84%, 81% and 94% for endometrioid carcinoma G3, in terms of F-measure, accuracy, precision and sensitivity, respectively. Furthermore, the proposed deep learning framework outperforms four state-of-the-art benchmarked methods by a significant margin (p < 0.001) in terms of accuracy, precision, sensitivity and F-measure, respectively. Additionally, a run time analysis demonstrates that the proposed method achieves excellent quantitative results with high efficiency in AI inference time (1.03 seconds per slide), making the proposed framework viable for practical clinical usage. These results highlight the efficacy and efficiency of the proposed model to assess MSI status of EC directly from histopathological slides.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. Data information.
a TCGA age distribution. b TCGA race distribution. c TCGA morphology subtype distribution. d TCGA cohort data image diversity.
Fig. 2
Fig. 2. Comparison in efficacy and efficiency.
a Box plots of the quantitative evaluation results. b Run time analysis for MSI prediction in comparison of the proposed DL method with ClassicMIL, CLAM, TOAD and KAT.
Fig. 3
Fig. 3. Flowchart and network architecture of the proposed method.
a Flowchart of the proposed DL method. (i) WSIs in multi-resolution pyramid tile-based structure Qll=1L are fed into (ii) the proposed foreground patch selection (FPS) model to rapidly locate high-resolution foreground patches uid,L without marker regions annotated by medical experts. Then, (iii) the weakly supervised tumor-like tissue segmentation model Ψtumor applies to the selected foreground patches uid,L to further generate the tumor-like patch attention score ξqj′L. Next, (iv) the proposed iterative patch sampling (IPS) method samples representative patches qj′L with high attention score ξqj′L. Afterwards, (v) the individual patch probability γjd,L of the representative patch qj′L is obtained using InceptionV3 classifier, while (vi) the individual patch decision weight ωqj′L of the representative patch qj′L is computed. Subsequently, (vii) the proposed weighted softmax integrated decision (WSID) model produces a reliable and robust slide level probability γ′d. (viii) Finally, the MSI status prediction DMSId of the d-th patient is generated. b The detailed networks of the proposed DL method.
Fig. 4
Fig. 4. Workflow diagram of the proposed DL method.
Firstly, a WSI in multi-resolution pyramid tile-based structure is fed into the proposed foreground patch selection (FPS) model to rapidly locate high-resolution foreground patches without marker regions annotated by medical experts. Then, the modified fully convolutional network model is applied to the selected foreground patches to further generate the tumor-like patch attention scores. Next, the proposed iterative patch sampling (IPS) method samples representative patches with high attention scores. Afterwards, individual patch probabilities of the representative patches are obtained using InceptionV3 classifier, while the individual patch decision weights of the representative patches are computed. Subsequently, the proposed weighted softmax integrated decision (WSID) model produces a reliable and robust slide level probability. Finally, the MSI status prediction is generated.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Raglan O, et al. Risk factors for endometrial cancer: An umbrella review of the literature. Int. J. cancer. 2019;145:1719–1730. doi: 10.1002/ijc.31961. - DOI - PubMed
    1. Berek, J. S. et al. Figo staging of endometrial cancer: 2023. Int. J. Gynecol. Obstetrics162, 383–394 (2023). - PubMed
    1. Lax SF, Pizer ES, Ronnett BM, Kurman RJ. Comparison of estrogen and progesterone receptor, ki-67, and p53 immunoreactivity in uterine endometrioid carcinoma and endometrioid carcinoma with squamous, mucinous, secretory, and ciliated cell differentiation. Hum. Pathol. 1998;29:924–931. doi: 10.1016/S0046-8177(98)90197-6. - DOI - PubMed
    1. Bokhman JV. Two pathogenetic types of endometrial carcinoma. Gynecologic Oncol. 1983;15:10–17. doi: 10.1016/0090-8258(83)90111-7. - DOI - PubMed
    1. Voss MA, et al. Should grade 3 endometrioid endometrial carcinoma be considered a type 2 cancer-a clinical and pathological evaluation. Gynecologic Oncol. 2012;124:15–20. doi: 10.1016/j.ygyno.2011.07.030. - DOI - PubMed