Ovarian torsion is a common concern in girls presenting to emergency care with pelvic or abdominal pain. The diagnosis is challenging to make accurately and quickly, relying on a combination of physical exam, history and radiologic evaluation. Failure to establish the diagnosis in a timely fashion can result in irreversible ovarian ischemia with implications for future fertility. Ultrasound is the mainstay of evaluation for ovarian torsion in the pediatric population. However, even with a high index of suspicion, imaging features are not pathognomonic.
We sought to develop an algorithm to aid radiologists in diagnosing ovarian torsion using machine learning from sonographic features and to evaluate the frequency of each sonographic element.
Materials and methods
All pediatric patients treated for ovarian torsion at a quaternary pediatric hospital over an 11-year period were identified by both an internal radiology database and hospital-based International Statistical Classification of Diseases and Related Health Problems (ICD) code review. Inclusion criteria were surgical confirmation of ovarian torsion and available imaging. Patients were excluded if the diagnosis could not be confirmed, no imaging was available for review, the ovary was not identified by imaging, or torsion involved other adnexal structures but spared the ovary. Data collection included: patient age; laterality of torsion; bilateral ovarian volumes; torsed ovarian position, i.e. whether medialized with respect to the mid-uterine line; presence or absence of Doppler signal within the torsed ovary; visualization of peripheral follicles; and presence of a mass or cyst, and free peritoneal fluid. Subsequently, we evaluated a non-torsed control cohort from April 2015 to May 2016. This cohort consisted of sequential girls and young adults presenting to the emergency department with abdominopelvic symptoms concerning for ovarian torsion but who were ultimately diagnosed otherwise. These features were then fed into supervised machine learning systems to identify and develop viable decision algorithms. We divided data into training and validation sets and assessed algorithm performance using sub-sets of the validation set.
We identified 119 torsion-confirmed cases and 331 torsion-absent cases. Of the torsion-confirmed cases, significant imaging differences were evident for girls younger than 1 year; these girls were then excluded from analysis, and 99 pediatric patients older than 1 year were included in our study. Among these 99, all variables demonstrated statistically significant differences between the torsion-confirmed and torsion-absent groups with P-values <0.005. Using any single variable to identify torsion provided only modest detection performance, with areas under the curve (AUC) for medialization, peripheral follicles, and absence of Doppler flow of 0.76±0.16, 0.66±0.14 and 0.82±0.14, respectively. The best decision tree using a combination of variables yielded an AUC of 0.96±0.07 and required knowledge of the presence of intra-ovarian flow, peripheral follicles, the volume of both ovaries, and the presence of cysts or masses.
Based on the largest series of pediatric ovarian torsion in the literature to date, we quantified sonographic features and used machine learning to create an algorithm to identify the presence of ovarian torsion — an algorithm that performs better than simple approaches relying on single features. Although complex combinations using multiple-interaction models provide slightly better performance, a clinically pragmatic decision tree can be employed to detect torsion, providing sensitivity levels of 95±14% and specificity of 92±2%.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Price excludes VAT (USA)
Tax calculation will be finalised during checkout.
Servaes S, Zurakowski D, Laufer MR et al (2007) Sonographic findings of ovarian torsion in children. Pediatr Radiol 37:446–451
Phillips GS, Parisi MT, Chew FS (2011) Imaging diagnosis of right lower quadrant pain in children. AJR Am J Roentgenol 196:W527–W534
Sintim-Damoa A, Majmudar AS, Cohen HL, Parvey LS (2017) Pediatric ovarian torsion: spectrum of imaging findings. Radiographics 37:1892–1908
Cass DL (2005) Ovarian torsion. Semin Pediatr Surg 14:86–92
McCloskey K, Grover S, Vuillermin P, Babl FE (2013) Ovarian torsion among girls presenting with abdominal pain: a retrospective cohort study. Emerg Med J 30:e11
Rey-Bellet Gasser C, Gehri M, Joseph J-M, Pauchard J-Y (2016) Is it ovarian torsion? A systematic literature review and evaluation of prediction signs. Pediatr Emerg Care 32:256–261
Dasgupta R, Renaud E, Goldin AB et al (2018) Ovarian torsion in pediatric and adolescent patients: a systematic review. J Pediatr Surg 53:1387–1391
Rha SE, Byun JY, Jung SE et al (2002) CT and MR imaging features of adnexal torsion. Radiographics 22:283–294
Albayram F, Hamper UM (2001) Ovarian and adnexal torsion: spectrum of sonographic findings with pathologic correlation. J Ultrasound Med 20:1083–1089
Peña JE, Ufberg D, Cooney N, Denis AL (2000) Usefulness of Doppler sonography in the diagnosis of ovarian torsion. Fertil Steril 73:1047–1050
Chang HC, Bhatt S, Dogra VS (2008) Pearls and pitfalls in diagnosis of ovarian torsion. Radiographics 28:1355–1368
Hurh PJ, Meyer JS, Shaaban A (2002) Ultrasound of a torsed ovary: characteristic gray-scale appearance despite normal arterial and venous flow on Doppler. Pediatr Radiol 32:586–588
Kokoska ER, Keller MS, Weber TR (2000) Acute ovarian torsion in children. Am J Surg 180:462–465
Chang Y-J, Yan D-C, Kong M-S et al (2008) Adnexal torsion in children. Pediatr Emerg Care 24:534–537
Graif M, Itzchak Y (1988) Sonographic evaluation of ovarian torsion in childhood and adolescence. AJR Am J Roentgenol 150:647–649
Linam LE, Darolia R, Naffaa LN et al (2007) US findings of adnexal torsion in children and adolescents: size really does matter. Pediatr Radiol 37:1013–1019
Valsky DV, Esh-Broder E, Cohen SM et al (2010) Added value of the gray-scale whirlpool sign in the diagnosis of adnexal torsion. Ultrasound Obstet Gynecol 36:630–634
Lee EJ, Kwon HC, Joo HJ et al (1998) Diagnosis of ovarian torsion with color Doppler sonography: depiction of twisted vascular pedicle. J Ultrasound Med 17:83–89
Dobson AJ (1990) An introduction to generalized linear models. Chapman & Hall, London
Breiman L, Friedman J, Stone CJ, Olshen RA (1984) Classification and regression trees. Taylor & Francis, London
Loh W-Y (2002) Regression trees with unbiased variable selection and interaction detection. Stat Sin 12:361–386
Guthrie BD, Adler MD, Powell EC (2010) Incidence and trends of pediatric ovarian torsion hospitalizations in the United States, 2000-2006. Pediatrics 125:532–538
Ngo A-V, Otjen JP, Parisi MT et al (2015) Pediatric ovarian torsion: a pictorial review. Pediatr Radiol 45:1845–1855
Oltmann SC, Fischer A, Barber R et al (2009) Cannot exclude torsion — a 15-year review. J Pediatr Surg 44:1212–1217
Ledesma-Medina J, Towbin RB, Newman B (1992) Pediatric case of the day. Right ovarian torsion, amputation, and calcification. Radiographics 12:199–200
Beaunoyer M, Chapdelaine J, Bouchard S, Ouimet A (2004) Asynchronous bilateral ovarian torsion. J Pediatr Surg 39:746–749
Oltmann SC, Fischer A, Barber R et al (2010) Pediatric ovarian malignancy presenting as ovarian torsion: incidence and relevance. J Pediatr Surg 45:135–139
Otjen JP, Stanescu L, Goldin A, Parisi MT (2015) A normal ovary in an abnormal location: a case of torsion. J Clin Ultrasound JCU 43:578–580
Helvie MA, Silver TM (1989) Ovarian torsion: sonographic evaluation. J Clin Ultrasound 17:327–332
Harmon JC, Binkovitz LA, Stephens J (2009) Uterine position in adnexal torsion: specificity and sensitivity of ipsilateral deviation of the uterus. Pediatr Radiol 39:354–358
Conflicts of interest
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Otjen, J.P., Stanescu, A.L., Alessio, A.M. et al. Ovarian torsion: developing a machine-learned algorithm for diagnosis. Pediatr Radiol 50, 706–714 (2020). https://doi.org/10.1007/s00247-019-04601-3
- Machine learning