Computer-Based Quantitative Assessment of Esophageal Transport Function Using Fluoroscopic Image Analysis

Background. Access to an objective quantitative assessment of esophageal transport function (ETF) is significantly limited in clinical practice due to two main factors: the insufficient availability of specialized equipment in healthcare institutions and the lack of adequate analysis techniques. The development of approaches utilizing conventional fluoroscopy with a liquid barium sulfate suspension contrast combined with automated processing has the potential to significantly increase the accessibility and objectivity of metric evaluation for patients with conditions involving swallowing disorders and esophageal motility impairments.

Objective: to develop and validate a proprietary algorithm for computer-based quantitative analysis of fluoroscopic images obtained from conventional esophageal fluoroscopy for the assessment of ETF in patients with related disorders.

Material and methods. A prospective comparative study included 34 patients with dysphagia (mean age 52.1±9.2 years). Each patient underwent modified fluoroscopy (a 60-second lateral video recording of the first swallow of 15 ml of standard barium suspension, corresponding to the first stage of contrast) and dynamic esophageal scintigraphy (DES) with 99mTc-technephyte, which was used as the reference method for quantitative parameters of mean esophageal transit time, based on our own results and literature data on quantitative ETF assessment. Fluoroscopic DICOM sequences were processed using the developed computer-based quantitative analysis of fluoroscopic images (QAFI) to generate intensity-time curves in selected regions of interest. The agreement between QAFI and DES results was assessed using dynamic time warping (DTW), cross-correlation, Bland–Altman analysis, and the intraclass correlation coefficient (ICC). Results. QAFI demonstrated high concordance with DES: mean DTW distance 0,055±0.012; synchronous crosscorrelation 0.925±0.06; ICC 0.915 (95% confidence interval 0.886–0.941). Bland–Altman analysis revealed no systematic bias (mean difference –0.01; limits of agreement –0.079 to 0.059). The diagnostic characteristics of QAFI were high: the area under the ROC curve was 0.94, with sensitivity of 0.90 and specificity of 0.88. The method's reproducibility was also high: intra-operator ICC>0,98, inter-operator ICC>0,97.

Conclusion. The developed proprietary algorithm for computer-based QAFI obtained by conventional fluoroscopy enables the acquisition of quantitative ETF parameters, demonstrating excellent diagnostic characteristics and good reproducibility. The obtained data confirm that QAFI could be a promising and widely accessible tool for use in routine clinical practice, as it is based on widely available fluoroscopy. The implementation of QAFI in clinical practice will facilitate a more objective and reproducible quantitative assessment of ETF in patients with esophageal motility disorders, expanding the capabilities of metric diagnostics without resorting to expensive techniques.

1. Li N, Yang WL, Cai MH, et al. Burden of gastroesophageal reflux disease in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of disease study 2019. BMC Public Health. 2023; 23(1): 582. https://doi.org/10.1186/s12889-023-15272-z.

2. Lee K, Hong SP, Yoo IK, et al. Global trends in incidence and prevalence of achalasia, 1925–2021: a systematic review and meta-analysis. United European Gastroenterol J. 2024; 12(4): 504–15. https://doi.org/10.1002/ueg2.12555.

3. Dellon ES, Hirano I. Epidemiology and natural history of eosinophilic esophagitis. Gastroenterology. 2018; 154(2): 319–32. https://doi.org/10.1053/j.gastro.2017.06.067.

4. Assadian M, Sanat ZM, Soleimani HA, Mikaeli J. Prevalence of different types of primary esophageal motility disorders and their associated factors in patients referring to Shariati Hospital during 2018–2019. Middle East J Dig Dis. 2022; 14(1): 70–6. https://doi.org/10.34172/mejdd.2022.258.

5. Carlson DA, Pandolfino JE. High-resolution manometry in clinical practice. Gastroenterol Hepatol. 2015; 11(6): 374–84.

6. Bashirov RA, Samoilenko LE, Ryzhkin SA, et al. Radionuclide diagnosis of esophageal dysmotility and gastroesophageal reflux in patients with systemic sclerosis. Journal of Radiology and Nuclear Medicine. 2023; 104(2): 124–37 (in Russ). https://doi.org/10.20862/0042-4676-2023-104-2-124-137.

7. O'Connor MK, Byrne PJ, Keeling P, Hennessy TP. Esophageal scintigraphy: applications and limitations in the study of esophageal disorders. Eur J Nucl Med. 1988; 14(3): 131–6. https://doi.org/10.1007/BF00293536.

8. Halder S, Acharya S, Kou W, et al. Mechanics informed fluoroscopy of esophageal transport. Biomech Model Mechanobiol. 2021; 20(3): 925–40. https://doi.org/10.1007/s10237-021-01420-0.

9. Scharitzer M, Pokieser P, Ekberg O. Oesophageal fluoroscopy in adults – when and why? Br J Radiol. 2024; 97(1159): 1222–33. https://doi.org/10.1093/bjr/tqae062.