Introduction
Anorectal disorders, causing fecal incontinence or evacuation dysfunction are common, affecting up to 5% of the general population (1). Most of these disorders result from an intricate combination of disturbances, including neurologic and pelvic floor musculature dysfunction, and have a significant impact in the quality of life of patients (2,3). Anorectal manometry (ARM) is the gold standard for the evaluation of anorectal functional disorders. Nevertheless, the accessibility to this exam is limited, and the complexity of data analysis and report is a significant drawback (4).
Aims & Methods
This pilot study aimed to develop and validate an artificial intelligence (AI) model to automatically differentiate motility patterns of fecal incontinence (FI) from obstructed defecation (OD) using ARM data.
We included ARM exams performed at a large gastrointestinal motility clinic in Brazil (Pelvia – Gastrointestinal Motility and Continence, Paraná, Brazil) between 2015 and 2021. All ARM procedures were performed using water-perfused probes with eight radial transducers with a distal latex balloon. We developed and tested multiple machine learning algorithms for the automatic interpretation of ARM data. Four models were tested: k-nearest neighbors (KNN), support vector machines (SVM), random forests (RF) and gradient boosting (xGB). These models were trained using a stratified 5-fold strategy. Their performance was assessed after fine-tuning of each model’s hyperparameters, using 90% of data for training and 10% of data for testing.
Results
A total of 827 ARM exams were used in this study, 493 of patients diagnosed with obstructed defecation (OD), while the remaining 334 presented FI. The performance marks of the different tuned models, including the precision, sensitivity and the F1-score, is summarized on table 1. After fine-tuning, the xGB model presented an overall accuracy (84.6 ± 2.9%), similar to that of RF (82.7 ± 4.8%) and SVM (81.0 ± 8.0%), and higher that of KNN (74.4 ± 3.8%). The xGB models showed the highest discriminating performance between OD and FI, with an area under the curve of 0.939.
| Model | Class | Precision
(mean % ± SD) | Sensitivity
(mean % ± SD) | F1-score
(mean % ± SD) |
| KNN (K-Nearest Neighbors) | OD | 79.6 ± 2.7
| 76.6 ± 5.4 | 78.0 ± 3.7
|
| FI | 68.0 ± 5.1
| 72.0 ± 3.7
| 69.8 ± 3.6
|
| SVM (Support Vector Machines) | OD | 85.0 ± 6.1
| 82.2 ± 8.8
| 83.6 ± 7.4
|
| FI | 75.6 ± 10.2
| 79.4 ± 8.0
| 77.6 ± 8.9
|
| RF (Random Forests) | OD | 81.8 ± 1.6
| 90.0 ± 7.3
| 85.6 ± 4.4
|
| FI | 84.0 ± 10.5
| 71.6 ± 1.3
| 76.8 ± 4.7
|
| xGB (Gradient Boosts) | OD | 83.6 ± 4.3
| 92.4 ± 3.6
| 87.6 ± 2.2
|
| FI | 87.4 ± 4.8
| 73.6 ± 8.2
| 79.6 ± 4.6
|
Conclusion
The tested machine learning algorithms, particularly the xGB model, accurately differentiated between FI and OD manometric patterns. Subsequent development of these tools may optimize the access to ARM studies, which may significantly impact on the management of patients with anorectal functional diseases.
References
1. Whitehead WE, Wald A, Diamant NE, et al. Functional disorders of the anus and rectum. Gut 1999;45 Suppl 2:Ii55-9.
2. Maeda Y, Vaizey CJ, Hollington P, et al. Physiological, psychological and behavioural characteristics of men and women with faecal incontinence. Colorectal Dis 2009;11:927-32.
3. Remes-Troche JM, Rao SS. Defecation disorders: neuromuscular aspects and treatment. Curr Gastroenterol Rep 2006;8:291-9.
4. Brown HW, Guan W, Schmuhl NB, et al. If We Don't Ask, They Won't Tell: Screening for Urinary and Fecal Incontinence by Primary Care Providers. J Am Board Fam Med 2018;31:774-782.
