Correlation of histogram analysis of apparent diffusion coefficient with uterine cervical pathologic finding

Yuning Lin, Hui Li, Ziqian Chen, Ping Ni, Qun Zhong, Huijuan Huang, Kumar Sandrasegaran

Research output: Contribution to journalArticle

34 Scopus citations

Abstract

OBJECTIVE. The purpose of this study was to investigate the application of histogram analysis of apparent diffusion coefficient (ADC) in characterizing pathologic features of cervical cancer and benign cervical lesions. SUBJECTS AND METHODS. This prospective study was approved by the institutional review board, and written informed consent was obtained. Seventy-three patients with cervical cancer (33-69 years old; 35 patients with International Federation of Gynecology and Obstetrics stage IB cervical cancer) and 38 patients (38-61 years old) with normal cervix or cervical benign lesions (control group) were enrolled. All patients underwent 3-T diffusionweighted imaging (DWI) with b values of 0 and 800 s/mm2. ADC values of the entire tumor in the patient group and the whole cervix volume in the control group were assessed. Mean ADC, median ADC, 25th and 75th percentiles of ADC, skewness, and kurtosis were calculated. Histogram parameters were compared between different pathologic features, as well as between stage IB cervical cancer and control groups. RESULTS. Mean ADC, median ADC, and 25th percentile of ADC were significantly higher for adenocarcinoma (p = 0.021, 0.006, and 0.004, respectively), and skewness was significantly higher for squamous cell carcinoma (p = 0.011). Median ADC was statistically significantly higher for well or moderately differentiated tumors (p = 0.044), and skewness was statistically significantly higher for poorly differentiated tumors (p = 0.004). No statistically significant difference of ADC histogram was observed between lymphovascular space invasion subgroups. All histogram parameters differed significantly between stage IB cervical cancer and control groups (p <0.05). CONCLUSION. Distribution of ADCs characterized by histogram analysis may help to distinguish early-stage cervical cancer from normal cervix or cervical benign lesions and may be useful for evaluating the different pathologic features of cervical cancer.

Original languageEnglish (US)
Pages (from-to)1125-1131
Number of pages7
JournalAmerican Journal of Roentgenology
Volume204
Issue number5
DOIs
StatePublished - May 1 2015

Keywords

  • Apparent diffusion coefficient (ADC)
  • Cervical cancer
  • Diffusion-weighted imaging (DWI)
  • Histogram

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Medicine(all)

Fingerprint Dive into the research topics of 'Correlation of histogram analysis of apparent diffusion coefficient with uterine cervical pathologic finding'. Together they form a unique fingerprint.

  • Cite this