Manual and semiautomatic segmentation of bone sarcomas on MRI have high similarity

Dionísio, Fernando Carrasco Ferreira; Oliveira, Larissa Santos; Hernandes, M. A.; Engel, Edgard Eduard; Rangayyan, Rangaraj M.; Azevedo-Marques, Paulo Mazzoncini de; Nogueira-Barbosa, Marcello Henrique

doi:10.1590/1414-431x20198962

Cited by 18 publications

(11 citation statements)

References 36 publications

(42 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the marrow is hyperintense, the threshold applied to the high b -value images allows a cleaner separation from other tissues, and thus requires less manual editing. This makes semi-automatic segmentation of bone marrow particularly suitable for women, while additional post-processing would be required in men to achieve matching levels of segmentation similarity [ 38 ]. Shimming technique had a small but significant effect on DSC values.…”

Section: Discussionmentioning

confidence: 99%

Semi-Automated Segmentation of Bone Metastases from Whole-Body MRI: Reproducibility of Apparent Diffusion Coefficient Measurements

et al. 2021

View full text Add to dashboard Cite

Using semi-automated software simplifies quantitative analysis of the visible burden of disease on whole-body MRI diffusion-weighted images. To establish the intra- and inter-observer reproducibility of apparent diffusion coefficient (ADC) measures, we retrospectively analyzed data from 20 patients with bone metastases from breast (BCa; n = 10; aged 62.3 ± 14.8) or prostate cancer (PCa; n = 10; aged 67.4 ± 9.0) who had undergone examinations at two timepoints, before and after hormone-therapy. Four independent observers processed all images twice, first segmenting the entire skeleton on diffusion-weighted images, and then isolating bone metastases via ADC histogram thresholding (ADC: 650–1400 µm2/s). Dice Similarity, Bland-Altman method, and Intraclass Correlation Coefficient were used to assess reproducibility. Inter-observer Dice similarity was moderate (0.71) for women with BCa and poor (0.40) for men with PCa. Nonetheless, the limits of agreement of the mean ADC were just ±6% for women with BCa and ±10% for men with PCa (mean ADCs: 941 and 999 µm2/s, respectively). Inter-observer Intraclass Correlation Coefficients of the ADC histogram parameters were consistently greater in women with BCa than in men with PCa. While scope remains for improving consistency of the volume segmented, the observer-dependent variability measured in this study was appropriate to distinguish the clinically meaningful changes of ADC observed in patients responding to therapy, as changes of at least 25% are of interest.

show abstract

Section: Discussionmentioning

confidence: 99%

Semi-Automated Segmentation of Bone Metastases from Whole-Body MRI: Reproducibility of Apparent Diffusion Coefficient Measurements

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Egger et al ( 19 , 20 ) demonstrated that semiautomatic segmentation would be faster than manual segmentation in glioblastomas and pituitary adenomas, which are tumors of the central nervous system. Dioní sio et al ( 23 ) described the time-saving advantage of the semiautomatic method over the manual method for the segmentation of bone sarcomas on MRI. We assume that our results did not reproduce those of studies of brain tumors and bone sarcomas, given the considerably greater tissue heterogeneity of the soft-tissue sarcomas included in our study sample; in the present study, semiautomatic segmentation methods tended to necessitate greater manual correction of the results of semiautomatic segmentation obtained by the software, as discussed in a previous study ( 32 ) .…”

Section: Discussionmentioning

confidence: 99%

“…Two of those radiologists (designated radiologist 1 and radiologist 2, respectively) were fellows in musculoskeletal radiology, and the third (designated radiologist 3) had five years of experience in the field. Manual segmentation was performed by manually delineating the tumor borders in each sectional slice, in all sectional slices in which tumor tissue was present ( Figure 2 ), according to procedures followed in previous studies ( 13 , 19 - 23 ) . Radiologist 1 manually segmented all cases again after a one-month interval, which allowed us to assess intraobserver variability.…”

Section: Methodsmentioning

confidence: 99%

Manual versus semiautomatic segmentation of soft-tissue sarcomas on magnetic resonance imaging: evaluation of similarity and comparison of segmentation times

et al. 2021

Self Cite

View full text Add to dashboard Cite

Objective: To evaluate the degree of similarity between manual and semiautomatic segmentation of soft-tissue sarcomas on magnetic resonance imaging (MRI). Materials and Methods: This was a retrospective study of 15 MRI examinations of patients with histopathologically confirmed soft-tissue sarcomas acquired before therapeutic intervention. Manual and semiautomatic segmentations were performed by three radiologists, working independently, using the software 3D Slicer. The Dice similarity coefficient (DSC) and the Hausdorff distance were calculated in order to evaluate the similarity between manual and semiautomatic segmentation. To compare the two modalities in terms of the tumor volumes obtained, we also calculated descriptive statistics and intraclass correlation coefficients (ICCs). Results: In the comparison between manual and semiautomatic segmentation, the DSC values ranged from 0.871 to 0.973. The comparison of the volumes segmented by the two modalities resulted in ICCs between 0.9927 and 0.9990. The DSC values ranged from 0.849 to 0.979 for intraobserver variability and from 0.741 to 0.972 for interobserver variability. There was no significant difference between the semiautomatic and manual modalities in terms of the segmentation times (p > 0.05). Conclusion: There appears to be a high degree of similarity between manual and semiautomatic segmentation, with no significant difference between the two modalities in terms of the time required for segmentation.

show abstract

“…The mean interobserver DSC between Observer-1 and Observer-2 was 0.93 ± 0.04 (maximum: 0.99; minimum: 0.78). The manual approach to segmentation is still considered equal or more reliable in comparison to the automated segmentation 14 .…”

Section: Resultsmentioning

confidence: 99%

An Adapted One-dimensional Computational Approach for Irregular ROI Analysis Improves Osteosarcoma Classification

Djuričić

Ahammer

Sopta

et al. 2021

Preprint

View full text Add to dashboard Cite

Background – The analysis of irregularly shaped tumour ROIs is hindered by the fact that most image analysis methods apart from first-order statistics are compatible only with rectangular ROIs. We thus aimed for the first implementation and testing of the 1-D MRI image analysis method that is fully compatible with irregular ROIs. Method – The retrospective prediction model of osteosarcoma chemoresponsiveness included T2-weighted MRI scans obtained before OsteoSa MAP neoadjuvant cytotoxic chemotherapy. Osteosarcoma morphology was quantified by calculating the one- and two-dimensional (1-D, 2-D) Higuchi dimensions (Dh), directionally and non-directionally. Results – The non-directional 1-D Dh reached a predictive AUC of 0.88, while the directional 1-D analysis along 180 radial lines robustly improved the predictive performance, reaching an AUC of 0.95, P<0.001 that is widely considered as nearly ideal. The optimal directional range was between 90o and 97o. Conclusions – We report the first validity testing of the 1-D analysis approach that is fully compatible with irregular ROIs. Such analytical adaptation to ROI shape in MRI has enhanced the osteosarcoma prediction performance over the previously reported standard 2-D analyses. The clinical importance of the early chemoresponsiveness prediction rests on its potential to prolong the survival of chemoresistant patients through personalised treatment adjustments.

show abstract

Manual and semiautomatic segmentation of bone sarcomas on MRI have high similarity

Cited by 18 publications

References 36 publications

Semi-Automated Segmentation of Bone Metastases from Whole-Body MRI: Reproducibility of Apparent Diffusion Coefficient Measurements

Semi-Automated Segmentation of Bone Metastases from Whole-Body MRI: Reproducibility of Apparent Diffusion Coefficient Measurements

Manual versus semiautomatic segmentation of soft-tissue sarcomas on magnetic resonance imaging: evaluation of similarity and comparison of segmentation times

An Adapted One-dimensional Computational Approach for Irregular ROI Analysis Improves Osteosarcoma Classification

Contact Info

Product

Resources

About