CT-based morphologic and radiomics features for the classification of MYCN gene amplification status in pediatric neuroblastoma

Tan, Eelin; Merchant, Khurshid; Kn, Bhanu Prakash; Cs, Arvind; Zhao, Joseph J; Saffari, Seyed Ehsan; Tan, Poh Hwa; Tang, Phua Hwee

doi:10.1007/s00381-022-05534-3

Cited by 14 publications

(12 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Methodsmentioning

confidence: 99%

“…Patients were scanned at multiple sites but treated at a single tertiary institution. The study cohort and imaging parameters were a subset of the previous research work 15 . The median volumetric CT dose index was 9.9 mGy (range, 2.1–19.8) based on a 16-cm–diameter phantom and 2.5 mGy (range, 0.4–11.3) based on a 32-cm–diameter phantom.…”

Section: Methodsmentioning

confidence: 99%

“…In our previous work, manual segmentation of the tumor region was performed using the open source 3-dimensional slicer medical image computing software, which was time consuming and laborious 15,32 . A modified U-Net model was trained for automatic segmentation of NBL tumors.…”

Section: Methodsmentioning

confidence: 99%

“…A subset of anonymized clinical and scan data of all pediatric patients diagnosed and treated for NBL at a tertiary pediatric hospital were identified through tumor board listings and text search of the Radiology Information System, from 2009 to 2020. 15 The inclusion criteria of the scans were pediatric patients with biopsy-proven NBL, known MYCN gene amplification status determined by fluorescence in situ hybridization, and pretreatment CECT. 31 The exclusion criteria were subject with scans registered only on a single scanner.…”

Section: Study Cohortmentioning

confidence: 99%

“…14 Moreover, only a small number of ML classifiers were evaluated. A more recent study by Tan et al 15 (2022) reported good predictive performance of a supervised ML model using contrast-enhanced CT (CECT) features to classify MYCN amplification in 54 patients. However, as the CT examinations were performed with different scanners at multiple sites, there was significant variability in the imaging data obtained, requiring correction.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Prediction of MYCN Gene Amplification in Pediatric Neuroblastomas: Development of a Deep Learning–Based Tool for Automatic Tumor Segmentation and Comparative Analysis of Computed Tomography–Based Radiomics Features Harmonization

Yeow

Teh

et al. 2023

J Comput Assist Tomogr

Self Cite

View full text Add to dashboard Cite

Objective MYCN oncogene amplification is closely linked to high-grade neuroblastoma with poor prognosis. Accurate quantification is essential for risk assessment, which guides clinical decision making and disease management. This study proposes an end-to-end deep-learning framework for automatic tumor segmentation of pediatric neuroblastomas and radiomics features-based classification of MYCN gene amplification. Methods Data from pretreatment contrast-enhanced computed tomography scans and MYCN status from 47 cases of pediatric neuroblastomas treated at a tertiary children's hospital from 2009 to 2020 were reviewed. Automated tumor segmentation and grading pipeline includes (1) a modified U-Net for tumor segmentation; (2) extraction of radiomic textural features; (3) feature-based ComBat harmonization for removal of variabilities across scanners; (4) feature selection using 2 approaches, namely, (a) an ensemble approach and (b) stepwise forward-and-backward selection method using logistic regression classifier; and (5) radiomics features-based classification of MYCN gene amplification using machine learning classifiers. Results Median train/test Dice score for modified U-Net was 0.728/0.680. The top 3 features from the ensemble approach were neighborhood gray-tone difference matrix (NGTDM) busyness, NGTDM strength, and gray-level run-length matrix (GLRLM) low gray-level run emphasis, whereas those from the stepwise approach were GLRLM low gray-level run emphasis, GLRLM high gray-level run emphasis, and NGTDM coarseness. The top-performing tumor classification algorithm achieved a weighted F1 score of 97%, an area under the receiver operating characteristic curve of 96.9%, an accuracy of 96.97%, and a negative predictive value of 100%. Harmonization-based tumor classification improved the accuracy by 2% to 3% for all classifiers. Conclusion The proposed end-to-end framework achieved high accuracy for MYCN gene amplification status classification.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Study Cohortmentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations