Download PDF
CC BY-NC-ND 4.0 · J Neurol Surg B Skull Base 2023; 84(06): 548-559
DOI: 10.1055/a-1941-3618
Original Article

The State of Machine Learning in Outcomes Prediction of Transsphenoidal Surgery: A Systematic Review

Darrion B. Yang 
1   Carle Illinois College of Medicine, University of Illinois Urbana Champaign, Champaign, Illinois, United States
,
Alexander D. Smith
1   Carle Illinois College of Medicine, University of Illinois Urbana Champaign, Champaign, Illinois, United States
,
Emily J. Smith   *
1   Carle Illinois College of Medicine, University of Illinois Urbana Champaign, Champaign, Illinois, United States
,
Anant Naik    *
1   Carle Illinois College of Medicine, University of Illinois Urbana Champaign, Champaign, Illinois, United States
,
Mika Janbahan   *
1   Carle Illinois College of Medicine, University of Illinois Urbana Champaign, Champaign, Illinois, United States
,
Charee M. Thompson
2   Department of Communication, University of Illinois Urbana Champaign, Champaign, Illinois, United States
,
Lav R. Varshney
3   Department of Electrical and Computer Engineering, University of Illinois Urbana Champaign, Urbana, Illinois, United States
,
Wael Hassaneen 
1   Carle Illinois College of Medicine, University of Illinois Urbana Champaign, Champaign, Illinois, United States
4   Department of Neurosurgery, Carle Foundation Hospital, Urbana, Illinois, United States
› Author Affiliations
› Further Information
Also available at
   Permissions and Reprints

Abstract

The purpose of this analysis is to assess the use of machine learning (ML) algorithms in the prediction of postoperative outcomes, including complications, recurrence, and death in transsphenoidal surgery. Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we systematically reviewed all papers that used at least one ML algorithm to predict outcomes after transsphenoidal surgery. We searched Scopus, PubMed, and Web of Science databases for studies published prior to May 12, 2021. We identified 13 studies enrolling 5,048 patients. We extracted the general characteristics of each study; the sensitivity, specificity, area under the curve (AUC) of the ML models developed as well as the features identified as important by the ML models. We identified 12 studies with 5,048 patients that included ML algorithms for adenomas, three with 1807 patients specifically for acromegaly, and five with 2105 patients specifically for Cushing's disease. Nearly all were single-institution studies. The studies used a heterogeneous mix of ML algorithms and features to build predictive models. All papers reported an AUC greater than 0.7, which indicates clinical utility. ML algorithms have the potential to predict postoperative outcomes of transsphenoidal surgery and can improve patient care. Ensemble algorithms and neural networks were often top performers when compared with other ML algorithms. Biochemical and preoperative features were most likely to be selected as important by ML models. Inexplicability remains a challenge, but algorithms such as local interpretable model–agnostic explanation or Shapley value can increase explainability of ML algorithms. Our analysis shows that ML algorithms have the potential to greatly assist surgeons in clinical decision making.

Keywords

transsphenoidal surgery - pituitary adenomas - machine learning - artificial intelligence - Cushing's disease - acromegaly

* These authors contributed equally.


Supplementary Material



Publication History

Received: 30 December 2021

Accepted: 03 March 2022

Accepted Manuscript online:
12 September 2022

Article published online:
23 November 2022

© 2022. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution-NonDerivative-NonCommercial License, permitting copying and reproduction so long as the original work is given appropriate credit. Contents may not be used for commercial purposes, or adapted, remixed, transformed or built upon. (https://creativecommons.org/licenses/by-nc-nd/4.0/)

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany

 

  • References

  • 1 Lake MG, Krook LS, Cruz SV. Pituitary adenomas: an overview. Am Fam Physician 2013; 88 (05) 319-327
    PubMedGoogle Scholar
  • 2 Zubair A, Das JM. Transsphenoidal hypophysectomy. In: StatPearls. StatPearls Publishing; 2021. Accessed September 6, 2022 at: http://www.ncbi.nlm.nih.gov/books/NBK556142/
    Google Scholar
  • 3 Halvorsen H, Ramm-Pettersen J, Josefsen R. et al. Surgical complications after transsphenoidal microscopic and endoscopic surgery for pituitary adenoma: a consecutive series of 506 procedures. Acta Neurochir (Wien) 2014; 156 (03) 441-449
    CrossrefPubMedGoogle Scholar
  • 4 Charalampaki P, Ayyad A, Kockro RA, Perneczky A. Surgical complications after endoscopic transsphenoidal pituitary surgery. J Clin Neurosci 2009; 16 (06) 786-789
    CrossrefPubMedGoogle Scholar
  • 5 Araujo-Castro M, Pascual-Corrales E, Martínez San Millan J. et al. Multidisciplinary protocol of preoperative and surgical management of patients with pituitary tumors candidates to pituitary surgery. Ann Endocrinol (Paris) 2021; 82 (01) 20-29
    CrossrefPubMedGoogle Scholar
  • 6 Lobatto DJ, de Vries F, Zamanipoor Najafabadi AH. et al. Preoperative risk factors for postoperative complications in endoscopic pituitary surgery: a systematic review. Pituitary 2018; 21 (01) 84-97
    CrossrefPubMedGoogle Scholar
  • 7 Rajkomar A, Dean J, Kohane I. Machine learning in medicine. N Engl J Med 2019; 380 (14) 1347-1358
    CrossrefPubMedGoogle Scholar
  • 8 Sidey-Gibbons JAM, Sidey-Gibbons CJ. Machine learning in medicine: a practical introduction. BMC Med Res Methodol 2019; 19 (01) 64
    CrossrefPubMedGoogle Scholar
  • 9 Berry MW, Mohamed A, Yap BW. Eds. Supervised and Unsupervised Learning for Data Science. Springer International Publishing; 2020
    CrossrefGoogle Scholar
  • 10 Soldozy S, Farzad F, Young S. et al. Pituitary tumors in the computational era: exploring novel approaches to diagnosis, and outcome prediction with machine learning. World Neurosurg 2021; 146: 315-321.e1
    CrossrefPubMedGoogle Scholar
  • 11 Staartjes VE, Zattra CM, Akeret K. et al. Neural network-based identification of patients at high risk for intraoperative cerebrospinal fluid leaks in endoscopic pituitary surgery. J Neurosurg 2019; 133 (02) 1-7
    PubMedGoogle Scholar
  • 12 Fan Y, Li Y, Li Y. et al. Development and assessment of machine learning algorithms for predicting remission after transsphenoidal surgery among patients with acromegaly. Endocrine 2020; 67 (02) 412-422
    CrossrefPubMedGoogle Scholar
  • 13 Voglis S, van Niftrik CHB, Staartjes VE. et al. Feasibility of machine learning based predictive modelling of postoperative hyponatremia after pituitary surgery. Pituitary 2020; 23 (05) 543-551
    CrossrefPubMedGoogle Scholar
  • 14 Hollon TC, Parikh A, Pandian B. et al. A machine learning approach to predict early outcomes after pituitary adenoma surgery. Neurosurg Focus 2018; 45 (05) E8
    CrossrefPubMedGoogle Scholar
  • 15 Shahrestani S, Cardinal T, Micko A. et al. Neural network modeling for prediction of recurrence, progression, and hormonal non-remission in patients following resection of functional pituitary adenomas. Pituitary 2021; 24 (04) 523-529
    CrossrefPubMedGoogle Scholar
  • 16 Machado LF, Elias PCL, Moreira AC, Dos Santos AC, Murta Junior LO. MRI radiomics for the prediction of recurrence in patients with clinically non-functioning pituitary macroadenomas. Comput Biol Med 2020; 124: 103966
    CrossrefPubMedGoogle Scholar
  • 17 Zhang W, Sun M, Fan Y. et al. Machine learning in preoperative prediction of postoperative immediate remission of histology-positive Cushing's disease. Front Endocrinol (Lausanne) 2021; 12: 635795
    CrossrefPubMedGoogle Scholar
  • 18 Zoli M, Staartjes VE, Guaraldi F. et al. Machine learning-based prediction of outcomes of the endoscopic endonasal approach in Cushing disease: is the future coming?. Neurosurg Focus 2020; 48 (06) E5
    CrossrefPubMedGoogle Scholar
  • 19 Fan Y, Li Y, Bao X. et al. Development of machine learning models for predicting postoperative delayed remission in patients with Cushing's disease. J Clin Endocrinol Metab 2021; a 106 (01) e217-e231
    CrossrefPubMedGoogle Scholar
  • 20 Liu Y, Liu X, Hong X. et al. Prediction of recurrence after transsphenoidal surgery for Cushing's disease: The use of machine learning algorithms. Neuroendocrinology 2019; 108 (03) 201-210
    CrossrefPubMedGoogle Scholar
  • 21 Fan Y, Li D, Liu Y, Feng M, Chen Q, Wang R. Toward better prediction of recurrence for Cushing's disease: a factorization-machine based neural approach. Int J Mach Learn Cybern 2021; b 12 (03) 625-633
    CrossrefPubMedGoogle Scholar
  • 22 Dai C, Fan Y, Li Y. et al. Development and interpretation of multiple machine learning models for predicting postoperative delayed remission of acromegaly patients during long-term follow-up. Front Endocrinol (Lausanne) 2020; 11: 643
    CrossrefPubMedGoogle Scholar
  • 23 Qiao N, Shen M, He W. et al. Machine learning in predicting early remission in patients after surgical treatment of acromegaly: a multicenter study. Pituitary 2021; 24 (01) 53-61
    CrossrefPubMedGoogle Scholar
  • 24 Villwock JA, Villwock MR, Goyal P, Deshaies EM. Current trends in surgical approach and outcomes following pituitary tumor resection. Laryngoscope 2015; 125 (06) 1307-1312
    CrossrefPubMedGoogle Scholar
  • 25 Mandrekar JN. Receiver operating characteristic curve in diagnostic test assessment. J Thorac Oncol 2010; 5 (09) 1315-1316
    CrossrefPubMedGoogle Scholar
  • 26 Acharya P, Mathur M. Artificial intelligence in dermatology: the ‘unsupervised’ learning. Br J Dermatol 2020; 182 (06) 1507-1508
    CrossrefPubMedGoogle Scholar
  • 27 Roohi A, Faust K, Djuric U, Diamandis P. Unsupervised machine learning in pathology: the next frontier. Surg Pathol Clin 2020; 13 (02) 349-358
    CrossrefPubMedGoogle Scholar
  • 28 Esteva A, Robicquet A, Ramsundar B. et al. A guide to deep learning in healthcare. Nat Med 2019; 25 (01) 24-29
    CrossrefPubMedGoogle Scholar
  • 29 Dinh A, Miertschin S, Young A, Mohanty SD. A data-driven approach to predicting diabetes and cardiovascular disease with machine learning. BMC Med Inform Decis Mak 2019; 19 (01) 211
    CrossrefPubMedGoogle Scholar
  • 30 Zhao Y, Wang T, Bove R. et al; SUMMIT Investigators. Ensemble learning predicts multiple sclerosis disease course in the SUMMIT study. NPJ Digit Med 2020; 3: 135
    CrossrefPubMedGoogle Scholar
  • 31 Ye S, Zhang H, Shi F, Guo J, Wang S, Zhang B. Ensemble learning to improve the prediction of fetal macrosomia and large-for-gestational age. J Clin Med 2020; 9 (02) E380
    CrossrefPubMedGoogle Scholar
  • 32 Bergquist T, Schaffter T, Yan Y. et al. Evaluation of crowdsourced mortality prediction models as a framework for assessing AI in medicine. 2021
    PubMedGoogle Scholar
  • 33 Hosni M, Abnane I, Idri A, Carrillo de Gea JM, Fernández Alemán JL. Reviewing ensemble classification methods in breast cancer. Comput Methods Programs Biomed 2019; 177: 89-112
    CrossrefPubMedGoogle Scholar
  • 34 Doupe P, Faghmous J, Basu S. Machine learning for health services researchers. Value Health 2019; 22 (07) 808-815
    CrossrefPubMedGoogle Scholar
  • 35 Natekin A, Knoll A. Gradient boosting machines, a tutorial. Front Neurorobot 2013; 7: 21
    CrossrefPubMedGoogle Scholar
  • 36 XGBoost Documentation—xgboost 1.6.0-dev documentation. Accessed September 6, 2022, at: https://xgboost.readthedocs.io/en/latest/
    PubMed
  • 37 Cruz JA, Wishart DS. Applications of machine learning in cancer prediction and prognosis. Cancer Inform 2007; 2: 59-77
    PubMedGoogle Scholar
  • 38 Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med 2019; 25 (01) 44-56
    CrossrefPubMedGoogle Scholar
  • 39 Jiang F, Jiang Y, Zhi H. et al. Artificial intelligence in healthcare: past, present and future. Stroke Vasc Neurol 2017; 2 (04) 230-243
    CrossrefPubMedGoogle Scholar
  • 40 Christodoulou E, Ma J, Collins GS, Steyerberg EW, Verbakel JY, Van Calster B. A systematic review shows no performance benefit of machine learning over logistic regression for clinical prediction models. J Clin Epidemiol 2019; 110: 12-22
    CrossrefPubMedGoogle Scholar
  • 41 Boulesteix A-L, Schmid M. Machine learning versus statistical modeling. Biom J 2014; 56 (04) 588-593
    CrossrefPubMedGoogle Scholar
  • 42 Selbst AD, Barocas S. The intuitive appeal of explainable machines. SSRN Journal 2018; DOI: 10.2139/ssrn.3126971.
    CrossrefPubMedGoogle Scholar
  • 43 Woods C, Thompson CJ. Risk of diabetes insipidus after pituitary surgery. Expert Rev Endocrinol Metab 2008; 3 (01) 23-27
    CrossrefPubMedGoogle Scholar
  • 44 Ironside N, Chatain G, Asuzu D. et al. Earlier post-operative hypocortisolemia may predict durable remission from Cushing's disease. Eur J Endocrinol 2018; 178 (03) 255-263
    CrossrefPubMedGoogle Scholar
  • 45 Bansal P, Lila A, Goroshi M. et al. Duration of post-operative hypocortisolism predicts sustained remission after pituitary surgery for Cushing's disease. Endocr Connect 2017; 6 (08) 625-636
    CrossrefPubMedGoogle Scholar
  • 46 Wang YY, Waqar M, Abou-Zeid A. et al. Value of early post-operative growth hormone testing in predicting long-term remission and residual disease after transsphenoidal surgery for acromegaly. Neuroendocrinology 2022; 112 (04) 345-357
    CrossrefPubMedGoogle Scholar
  • 47 Butenschoen VM, von Werder A, Bette S. et al. Transsphenoidal pituitary adenoma resection: do early post-operative cortisol levels predict permanent long-term hypocortisolism?. Neurosurg Rev 2022; 45 (02) 1353-1362
    CrossrefPubMedGoogle Scholar
  • 48 Lu C, Lin XT, Yang DT, Liu YC, He WT, Zhong XL. Pre- and post-operative hypothalamic-pituitary-thyroidal axis function in patients with prolactinoma, growth hormone tumour and ACTH tumour. Chin Med J (Engl) 1989; 102 (04) 306-312
    PubMedGoogle Scholar
  • 49 Agrawal N, Ioachimescu AG. Prognostic factors of biochemical remission after transsphenoidal surgery for acromegaly: a structured review. Pituitary 2020; 23 (05) 582-594
    CrossrefPubMedGoogle Scholar
  • 50 Shirvani M, Motiei-Langroudi R, Sadeghian H. Outcome of microscopic transsphenoidal surgery in Cushing disease: a case series of 96 patients. World Neurosurg 2016; 87: 170-175
    CrossrefPubMedGoogle Scholar
  • 51 Bourdelot A, Coste J, Hazebroucq V. et al. Clinical, hormonal and magnetic resonance imaging (MRI) predictors of transsphenoidal surgery outcome in acromegaly. Eur J Endocrinol 2004; 150 (06) 763-771
    CrossrefPubMedGoogle Scholar
  • 52 Patel PN, Stafford AM, Patrinely JR. et al. Risk factors for intraoperative and postoperative cerebrospinal fluid leaks in endoscopic transsphenoidal sellar surgery. Otolaryngol Head Neck Surg 2018; 158 (05) 952-960
    CrossrefPubMedGoogle Scholar
  • 53 Hussain NS, Piper M, Ludlam WG, Ludlam WH, Fuller CJ, Mayberg MR. Delayed postoperative hyponatremia after transsphenoidal surgery: prevalence and associated factors. J Neurosurg 2013; 119 (06) 1453-1460
    CrossrefPubMedGoogle Scholar
  • 54 Braileanu M, Hu R, Hoch MJ. et al. Pre-operative MRI predictors of hormonal remission status post pituitary adenoma resection. Clin Imaging 2019; 55: 29-34
    CrossrefPubMedGoogle Scholar