Dataset Information

Decoding intra-tumoral spatial heterogeneity on radiological images using the Hilbert curve

ABSTRACT:

Background

Current intra-tumoral heterogeneous feature extraction in radiology is limited to the use of a single slice or the region of interest within a few context-associated slices, and the decoding of intra-tumoral spatial heterogeneity using whole tumor samples is rare. We aim to propose a mathematical model of space-filling curve-based spatial correspondence mapping to interpret intra-tumoral spatial locality and heterogeneity.

Methods

A Hilbert curve-based approach was employed to decode and visualize intra-tumoral spatial heterogeneity by expanding the tumor volume to a two-dimensional (2D) matrix in voxels while preserving the spatial locality of the neighboring voxels. The proposed method was validated using three-dimensional (3D) volumes constructed from lung nodules from the LIDC-IDRI dataset, regular axial plane images, and 3D blocks.

Results

Dimensionality reduction of the Hilbert volume with a single regular axial plane image showed a sparse and scattered pixel distribution on the corresponding 2D matrix. However, for 3D blocks and lung tumor inside the volume, the dimensionality reduction to the 2D matrix indicated regular and concentrated squares and rectangles. For classification into benign and malignant masses using lung nodules from the LIDC-IDRI dataset, the Inception-V4 indicated that the Hilbert matrix images improved accuracy (85.54% vs. 73.22%, p < 0.001) compared to the original CT images of the test dataset.

Conclusions

Our study indicates that Hilbert curve-based spatial correspondence mapping is promising for decoding intra-tumoral spatial heterogeneity of partial or whole tumor samples on radiological images. This spatial-locality-preserving approach for voxel expansion enables existing radiomics and convolution neural networks to filter structured and spatially correlated high-dimensional intra-tumoral heterogeneity.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13244-021-01100-8.

SUBMITTER: Wang L

PROVIDER: S-EPMC8557226 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Intra-individual tumoral heterogeneity (ITH) is a hallmark of solid tumors and impedes accurate genomic diagnosis and selection of proper therapy. The purpose of this study was to identify ITH of ovarian serous adenocarcinomas (OSAs) and to determine the utility of ascitic cancer cells as a resource for mutation profiling in spite of ITH. We performed whole-exome sequencing, copy number profiling, and DNA methylation profiling of four OSA genomes using multiregional biopsies from 13 intraovarian lesions, 12 extraovarian tumor lesions (omentum/peritoneum), and ascitic cells. We observed substantial levels of heterogeneity in mutations and copy number alterations (CNAs) of the OSAs. We categorized the mutations into 'common', 'shared' and 'private' according to the regional distribution. Six common, 8 shared, and 24 private mutations were observed in known cancer-related genes,. but common mutations had a higher mutant allele frequency and included TP53 mutations in all four OSAs. Region-specific chromosomal amplifications and deletions involving BRCA1, PIK3CA, and RB1 were also identified. Of note, the mutations detected in ascitic cancer cells represented 92.3-100% of overall somatic mutations in the given case. Phylogenetic analyses of ascitic genomes predicted a polyseeding origin of somatic mutations in ascitic cells. Our results demonstrate that despite ITH, somatic mutations, CNAs, and DNA methylations in both â??commonâ?? category and cancer-related genes were highly conserved in ascitic cells of OSAs, highlighting the clinical relevance of genome analysis of ascitic cells. Ascitic tumor cells may serve as a potential resource to discover somatic mutations of primary OSA with diagnostic and therapeutic relevance. Genome wide DNA methylation profiling of ascitic cells as well as biopsies from ovarian serous adenocarcinomas cases obtained by Illumina Infinium 450k Human DNA methylation Beadchip Bisulphite converted DNA from the 16 samples were hybridized to the Illumina Infinium 450k Human Methylation Beadchip