Applying artificial intelligence to longitudinal imaging analysis of vestibular schwannoma following radiosurgery

Abstract

Artificial intelligence (AI) has been applied with considerable success in the fields of radiology, pathology, and neurosurgery. It is expected that AI will soon be used to optimize strategies for the clinical management of patients based on intensive imaging follow-up. Our objective in this study was to establish an algorithm by which to automate the volumetric measurement of vestibular schwannoma (VS) using a series of parametric MR images following radiosurgery. Based on a sample of 861 consecutive patients who underwent Gamma Knife radiosurgery (GKRS) between 1993 and 2008, the proposed end-to-end deep-learning scheme with automated pre-processing pipeline was applied to a series of 1290 MR examinations (T1W+C, and T2W parametric MR images). All of which were performed under consistent imaging acquisition protocols. The relative volume difference (RVD) between AI-based volumetric measurements and clinical measurements performed by expert radiologists were + 1.74%, − 0.31%, − 0.44%, − 0.19%, − 0.01%, and + 0.26% at each follow-up time point, regardless of the state of the tumor (progressed, pseudo-progressed, or regressed). This study outlines an approach to the evaluation of treatment responses via novel volumetric measurement algorithm, and can be used longitudinally following GKRS for VS. The proposed deep learning AI scheme is applicable to longitudinal follow-up assessments following a variety of therapeutic interventions.

Convolutional Neural Networks for Classifying Laterality of Vestibular Schwannomas on Single MRI Slices—A Feasibility Study

“…One publication used an NN to predict vs. recurrence following surgery from clinical parameters in tabular format [23]. The remaining four publications used NNs to segment VS for either radiotherapy planning or response assessment [24][25][26][27].…”

Application of artificial intelligence in the otoneurology diagnostic process – an overview

“…(2021) zaproponowali kompleksową metodę segmentacji opartej na głębokim uczeniu się z wykorzystaniem wieloparametrycznych obrazów rezonansu magnetycznego. Zaprezentowany algorytm umożliwia nie tylko automatyzację analizy, w której dokładność modelu AI przekroczyła 99% przy pomiarze objętości guza, lecz także opisanie kierunku progresji, regresji lub pseudoregresji guza po radioterapii [22].…”

“…Wyniki badań z ostatnich lat są spójne w przypadku określania korelacji między zmianą objętości i charakterystyką obrazów [20][21][22]. Biorąc pod uwagę aspekt kliniczny, wielkość guza stanowi o zaawansowaniu choroby i nasileniu objawów, a więc jest to decydujący aspekt przed podjęciem leczenia chirurgicznego.…”

“…Ponadto opracowano system ML do przewidywania u pacjentów z nerwiakiem wyników zachowania słyszenia, których średnia dokładność wynosi 62-90% w zależności od przyjętej metody [23]. Efekty podjętego leczenia będą zweryfikowane przez analizę badań przy użyciu AI, na podstawie której można przewidzieć odpowiedź na leczenie (regresja, pseudoregresja) [22].…”

Contribution of spectroscopic magnetic resonance imaging to target volume delineation in Gamma Knife Radiosurgery: Myth or reality?

“…Moreover, some autopsy series of brain metastases confirmed infiltrative growth in radiologically healthy surrounding tissues in more than 60% of cases. This unseen malignant component is responsible of 80 % of "early recurrence" which should be considered as natural evolution of the main tumor [5]. In the management of high grade gliomas, the radiosurgeons are faced either to carcinologic incomplete procedures or to overestimated target irradiation with unbalanced benefit/risk action mostly related to radiation-induced brain necrosis [6].…”