Volumetric glioma quantification: comparison of manual and semi-automatic tumor segmentation for the quantification of tumor growth

Abstract

Manual segmentation of HGG is very time-consuming and using the SAM could increase the efficiency of this process. However, the accuracy of the SAM ultimately depends on the expert doing the editing. Our study confirmed a considerable inter-observer variability among experts defining tumor volume from volumetric MRIs.

Automated brain tumour detection and segmentation using superpixel-based extremely randomized trees in FLAIR MRI

“…In clinical studies, current Response Assessment in Neurooncology (RANO) criterion simply uses a bidirectional measurement to determine tumour volume for assessing treatment response [4]. Although a full 3D volume measurement may provide a more accurate volume assessment, there is a need for accurate and fully automated methods since manual segmentation (region of interest drawing) around tumour margins on a slice-by-slice basis is time-consuming and can take 12 min or more per tumour, with semiautomatic methods taking 3–5 min [5, 6]. T2w/FLAIR images can also be useful to help define the target volumes for radiotherapy planning of high-grade gliomas [2, 5]; hence, an automated segmentation that is not subject to operator subjectivity may be beneficial [5].…”

Volumetric assessment of unaffected parenchyma and Wilms’ tumours: analysis of response to chemotherapy and surgery using a semi‐automated segmentation algorithm in children with renal neoplasms

“…This introduces the potential risk of error, but to a lesser degree and with less variability than previously reported with human interpretation of diagnostic images . Odland et al explored similar technology in volumetric glioma assessment using semi‐automatic segmentation and showed that the technology was accurate, and resulted in less variability in measurements than manual segmentation by experts; however, they acknowledge that the accuracy of this technology is dependent on the expert responsible for training the algorithm. Rana et al described a method of automatic segmentation for mandibular odontogenic cysts and tumours, and found that their technique for auto‐segmentation was as accurate as semi‐automatic and manual segmentation while being much faster to complete.…”

“…Currently, manual estimation of tumour volume using segmentation is somewhat cumbersome and may result in variability due to technique and individuals. To date, manual measurement of areas of interest in radiological images has been shown to be feasible in previous studies across several specialty areas .…”

Virtual Raters for Reproducible and Objective Assessments in Radiology

“…The proposed method represents a continuous hybrid between semi- and fully automatic segmentation methods and is able to combine their strengths. In contrast to existing semi-automatic methods 3 13 14 15 the knowledge gained during annotation of single data sets is conserved and is subsequently used to rate images of novel patients or other time points of the same patient. Using our tool in patient-specific semi-automatic mode, without taking prior knowledge into account, leads to ICC values comparable to previously published results 35 .…”

“…While automatic 10 11 12 and semi-automatic methods 3 13 14 15 exist, these are usually not employed in clinical routine for two major reasons. Firstly, physicians are usually pressed for time and available tools are not easily integrated in the standard clinical workflow.…”