Radiomics in assessing neurosonographic changes in newborns with diabetic fetopathy: analysis of ultrasound images

Introduction. Early noninvasive assessment of brain changes in newborns is a significant challenge in pediatrics. This article presents an approach to noninvasive assessment of brain changes in newborns using radiomics analysis of ultrasound images. Radiomics analysis allows characterization of the morphological structure of neurosonographic ultrasound images using a set of texture parameters and the identification of changes invisible to the naked eye.

Aim: To investigate the feasibility of using radiomics analysis of ultrasound images to detect brain changes in diabetic fetopathy in newborns.

Material and Methods. Data were collected from brain ultrasound images of 89 full-term neonates (gestational age greater than 37 weeks), including 45 (51%) healthy neonates (control group) and 44 (49%) with diabetic fetopathy (study group). Data were extracted using specialized projections to display four locations:

1. Frontal lobe (F0 scan at the level of the anterior sections of both frontal lobes): 45 healthy and 37 patients.
2. Parasagittal section in the choroid plexus area (S2 scan in the parasagittal plane): 41 healthy and 40 patients.
3. Sagittal section in the corpus callosum area (S0 scan in the midsagittal plane): 44 healthy and 40 patients.
4. Frontal section in the periventricular region (F4 scanning in the area of the parietal and temporal lobes, as well as the cerebellum): 45 healthy and 44 patients.

Results. When conducting B-mode neurosonography, the same frequency of subependymal cysts and lateral ventricular dilation was observed in both groups (7% vs. 5%; p = 0.53), but intraventricular hemorrhages and periventricular edema were observed only in the main group (7% vs. 0%; p < 0,05). As a result of radiomics analysis of ultrasound images of the brain, radiomics predictors of texture changes in newborns with diabetic fetopathy were established in four localizations. Classification models were built and ROC analysis was performed. The best results were shown by Model 1 for the frontal lobe (accuracy – 0.71, AUC = 0,69) and Model 4 for the periventricular region (accuracy – 0.89, AUC = 0,85). The established textural changes in the brain of newborns with diabetic fetopathy manifest as follows: an uneven, chaotic distribution of echogenicity with multiple hyperechoic areas is observed in the frontal lobe. The periventricular zone exhibits a marked, diffuse increase in echogenicity, creating a homogenized effect on the image. Radiomics analysis of ultrasound images can reveal changes in brain texture that are not detectable with standard neurosonography.

Conclusions. Multiparametric analysis of ultrasound images using a radiomics approach demonstrated the ability to detect structural changes in the brain of newborns with diabetic fetopathy. The results confirm the effectiveness of radiomics analysis in identifying subtle neuroanatomical changes.

1. Dedov I.I., Shestakova M.V., Vikulova O.K., Zheleznyakova A.V., Isakov M.A., Sazonova D.V., Mokrysheva N.G. Diabetes mellitus in the Russian Federation: dynamics of epidemiological indicators according to the Federal Register of Diabetes Mellitus for the period 2010–2022. Diabetes mellitus. 2023;26(2):104–123. https://doi.org/10.14341/DM13035

2. Holt R.I.G., Flyvbjerg A. (eds.). Textbook of diabetes. Sixth edition. Hoboken, NJ: Wiley-Blackwell, 2024. ISBN: 978-1-119-69742-8.

3. Riddle M.C., Cefalu W.T., Evans P.H., Gerstein H.C., Nauck M.A., Oh W.K. et al. Consensus Report: Definition and Interpretation of Remission in Type 2 Diabetes. Diabetes Care. 2021;44(10):2438–2444. https://doi.org/10.2337/dci21-0034

4. Lebedeva M.A. Features of pregnancy management and childbirth in diabetic fetopathy: dis. ... candidate of medical sciences: 14.01.01 / M.A. Lebedeva. [Place of protection: Federal State Budgetary Educational Institution of Higher Education “Moscow State University of Medicine and Tomatology named after A.I. Evdokimov”]. Moscow, 2021:120.

5. Herath M.P., Beckett J.M., Hills A.P., Byrne N.M., Ahuja K.D.K. Gestational diabetes mellitus and infant adiposity at birth: A systematic review and meta-analysis of therapeutic interventions. J. Clin. Med. 2021;10(4):835. https://doi.org/10.3390/jcm10040835

6. Nudnov N.V., Bit-Yunan E.V., Shakhvalieva E.S., Borisov A.A., Sultanova P.N., Ivannikov M.E.et al. Radiomics in the differential diagnosis of focal brain lesions: a retrospective study. Radiation diagnosis and therapy. 2024;15(3):32–38. https://doi.org/10.22328/2079-53432024-15-3-32-38

7. Shukhratbekova M.Kh. The role of neurosonography in the diagnosis of neuropsychic developmental delay in early children. Economy and Society. 2023;(10 (113)):1. URL: https://cyberleninka.ru/article/n/rol-neyrosonografii-v-diagnostike-zaderzhki-nervno-psihicheskogorazvitiya-u-detey-rannego-vozrasta (09.14.2025).

8. Hatt M., Le Rest C.C., Tixier F., Badic B., Schick U., Visvikis D. Radiomics: Data are also images. J. Nucl. Med. 2019;60(Suppl. 2):38S–44S. https://doi.org/10.2967/jnumed.118.220582

9. Lambin P., Rios-Velazquez E., Leijenaar R., Carvalho S., van Stiphout R.G., Granton P. et al. Radiomics: Extracting more information from medical images using advanced feature analysis. Eur. J. Cancer. 2012;48(4):441–446. https://doi.org/10.1016/j.ejca.2011.11.036

10. Jiang H., Chen L., Zhao Y.J., Lin Z.Y., Yang H. Machine learning-based ultrasomics for predicting subacromial impingement syndrome stages. J. Ultrasound Med. 2022;41(9):2279–2285. https://doi.org/10.1002/jum.15914

11. Jia Y., Yang J., Zhu Y., Nie F., Wu H., Duan Y., Chen K. Ultrasoundbased radiomics: current status, challenges and future opportunities. Med. Ultrason. 2022;24(4):451–460. https://doi.org/10.11152/mu-3248

12. McCague C., Ramlee S., Reinius M., Selby I., Hulse D., Piyatissa P. et al. Introduction to radiomics for a clinical audience. Clin. Radiol. 2023;78(2):83–98. https://doi.org/10.1016/j.crad.2022.08.149

13. Van Griethuysen J.J.M., Fedorov A., Parmar C., Hosny A., Aucoin N., Narayan V. et al. Computational radiomics system to decode the radiographic phenotype. Cancer Research. 2017;77(21):e104–e107. https://doi.org/10.1158/0008-5472.CAN-17-0339

14. Triantafyllou M., Vassalou E.E., Goulianou A.M., Tosounidis T.H., Marias K., Karantanas A.H. et al. The effect of ultrasound image preprocessing on radiomics feature quality: A study on shoulder ultrasound. J. Imaging Inform. Med. 2025 Feb 6. Epub ahead of print. https://doi.org/10.1007/s10278-025-01421-w

15. Zwanenburg A., Vallières M., Abdalah M.A., Aerts H.J.W.L., Andrearczyk V., Apte A. et al. The image biomarker standardization initiative: Standardized quantitative radiomics for high-throughput image-based phenotyping. Radiology. 2020;295(2):328–338. https://doi.org/10.1148/radiol.2020191145

16. Arakelian G.A., Orazmuradov A.A., Mayatskaya T.A., Bekbaeva I.V., Kotaysh G.A. Features of newborns from mothers with gestational diabetes mellitus and pregestational obesity. Obstetrics and Gynecology: news, opinions, training. 2020;8(3).Suppl.:24–29. https://doi.org/10.24411/2303-9698-2020-13904

17. Sakharova E.S., Keshishyan E.S., Alyamovskaya G.A. Relationship between the results of cranial ultrasound examination – neurosonography – and the outcomes of psychomotor development of children born extremely premature. Russian Bulletin of Perinatology and Pediatrics. 2019;64(5):33–37. https://doi.org/10.21508/1027-4065-2019-64-5-33-37

18. Sultan L.R., Ramirez-Suarez K.I., Schenkel S.R., Schaeubinger M.M., Cerron-Vela C., Kgole S.W. et al. Brain ultrasound radiomics identify textural differences in basal ganglia and white matter between full term newborns HIV-exposed uninfected and HIV-unexposed in Botswana. Early Hum. Dev. 2025;210:106368. https://doi.org/10.1016/j.earlhumdev.2025.106368