Exploratory Study of Apparent Diffusion Coefficient Histogram Metrics in Assessing Pancreatic Malignancy

Artificial Intelligence Applied to Pancreatic Imaging: A Narrative Review

The diagnosis, evaluation, and treatment planning of pancreatic pathologies usually require the combined use of different imaging modalities, mainly, computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET). Artificial intelligence (AI) has the potential to transform the clinical practice of medical imaging and has been applied to various radiological techniques for different purposes, such as segmentation, lesion detection, characterization, risk stratification, or prediction of response to treatments. The aim of the present narrative review is to assess the available literature on the role of AI applied to pancreatic imaging. Up to now, the use of computer-aided diagnosis (CAD) and radiomics in pancreatic imaging has proven to be useful for both non-oncological and oncological purposes and represents a promising tool for personalized approaches to patients. Although great developments have occurred in recent years, it is important to address the obstacles that still need to be overcome before these technologies can be implemented into our clinical routine, mainly considering the heterogeneity among studies.

Noise reduction in diffusion weighted MRI of the pancreas using an L1-regularized iterative SENSE reconstruction

OBJECTIVES

To prospectively evaluate an L1 regularized iterative SENSE reconstruction (L1-R SENSE) to eliminate band-like artifacts frequently seen with parallel imaging (SENSE) at high acceleration factors in high resolution diffusion weighted magnetic resonance imaging of the pancreas.

METHODS

Fourteen patients with pancreatic ductal adenocarcinoma (PDAC) underwent respiratory triggered DWI ss-EPI at a resolution of 2.5 × 2.5 × 3 mm3 with uniform undersampling in the phase encoding direction (AP axis) with an acceleration factor of 4. Data were reconstructed using the standard SENSE reconstruction routine of the vendor and an iterative SENSE reconstruction employing L1 regularization after a wavelet sparsifying transformation (L1-R SENSE). Retrospective reconstruction of the data with a lower number of averages was performed using both reconstruction methods. Two radiologists independently assessed noise artifacts, anatomical details and image quality (IQ) subjectively with a 4-point scale. Apparent diffusion coefficient (ADC) and covariance (CV) of ADC estimated from images reconstructed at a different number of averages for PDAC and the normal pancreas were assessed.

RESULTS

L1-R SENSE resulted in higher IQ and less noise artifacts than SENSE. Anatomical details were significantly higher for SENSE in one reader. Mean ADC of PDAC and normal pancreas were significantly higher for L1-R SENSE than SENSE. L1-R SENSE revealed lower CV of ADC for normal pancreas compared to SENSE, whereas no difference was noted for PDAC.

CONCLUSION

Compared with traditional SENSE reconstruction, L1-R SENSE effectively reduces band-like noise and improves the robustness of the ADC estimation from acquisitions using single-shot DW-EPI of the pancreas.

Accuracy of quantitative diffusion-weighted imaging for differentiating benign and malignant pancreatic lesions: a systematic review and meta-analysis

BACKGROUND

A variety of imaging techniques can be used to evaluate diffusion characteristics to differentiate malignant and benign pancreatic lesions. The diagnostic performance of diffusion parameters has not been systematic assessed.

PURPOSE

We aimed to investigate the diagnostic efficacy of quantitative diffusion-weighted imaging (DWI) for pancreatic lesions.

METHODS

A literature search was conducted using the PubMed, Embase, and Cochrane Library databases for studies from inception to March 30, 2020, which involves the quantitative diagnostic performance of diffusion-weighted imaging (DWI) and intravoxel incoherent motion (IVIM) in the pancreas. Studies were reviewed according to inclusion and exclusion criteria. The quality of articles was evaluated by the Quality Assessment of Diagnostic Accuracy Studies-2 (QUATAS-2). A bivariate random-effects model was used to evaluate pooled sensitivities and specificities. Univariable meta-regression analysis was used to test the effects of factors that contributed to the heterogeneity.

RESULTS

A total of 31 studies involving 1558 patients were ultimately eligible for data extraction. The lowest heterogeneity was found in specificity of perfusion fraction (f) with the I2 value was 17.97% and Cochran p value was 0.28. However, high heterogeneities were found for the other parameters (all I2 > 50%). There was no publication bias found in funnel plot (p = 0.30) for the apparent diffusion coefficient (ADC) parameter. The pooled sensitivities for ADC, f, pure diffusion coefficient (D), and pseudo diffusivity coefficient (D*) were 83%, 81%, 76%, and 84%, respectively. The pooled specificities for ADC, f, D, and D* were 87%, 83%, 69%, and 81% respectively. The areas under the curves for ADC, f, D, and D* were 0.92, 0.87, 0.79, and 0.87 respectively.

CONCLUSION

Quantitative DWI and IVIM have a good diagnostic performance for differentiating malignant and benign pancreatic lesions.

KEY POINTS

• IVIM has high sensitivity and specificity (84% and 83%, respectively) for differential diagnosis of pancreatic lesions, which is comparable to that of the ADC (83% and 87%, respectively). • The ADC has an excellent diagnostic performance for differentiating malignant from benign IPMNs (sensitivity, 0.83; specificity, 0.92); the f has the best diagnostic performance for differentiating pancreatic carcinoma from PNET (sensitivity, 0.85; specificity, 0.85). • For the ADC, using a maximal b value < 800 s/mm2 has a higher diagnostic accuracy than ≥ 800 s/mm2; performing in a high field strength (3.0 T) system has a higher diagnostic accuracy than a low field strength (1.5 T) for pancreatic lesions.

Diffusion-weighted imaging with histogram analysis of the apparent diffusion coefficient maps in the diagnosis of parotid tumours

The aim of this study was to investigate the role of diffusion-weighted imaging (DWI) with histogram analysis of apparent diffusion coefficient (ADC) maps in the characterization of parotid tumours. This prospective study included 39 patients with parotid tumours. All patients underwent magnetic resonance imaging with DWI, and ADC maps were generated. The whole lesion was selected to obtain histogram-related parameters, including the mean (ADC_mean), minimum (ADC_min), maximum (ADC_max), skewness, and kurtosis of the ADC. The final diagnosis included pleomorphic adenoma (PA; n=18), Warthin tumour (WT; n=12), and salivary gland malignancy (SGM; n=9). ADC_mean (×10^-3mm²/s) was 1.93±0.34 for PA, 1.01±0.11 for WT, and 1.26±0.54 for SGM. There was a significant difference in whole lesion ADC_mean among the three study groups. Skewness had the best diagnostic performance in differentiating PA from WT (P=0.001; best detected cut-off 0.41, area under the curve (AUC) 0.990) and in discriminating WT from SGM (P=0.03; best detected cut-off 0.74, AUC 0.806). The whole lesion ADC_mean value had best diagnostic performance in differentiating PA from SGM (P=0.007; best detected cut-off 1.16×10^-3mm²/s, AUC 0.948). In conclusion, histogram analysis of ADC maps may offer added value in the differentiation of parotid tumours.

Identification of a competing endogenous RNA network associated with prognosis of pancreatic adenocarcinoma

Background

Emerging evidence suggests that competing endogenous RNAs plays a crucial role in the development and progress of pancreatic adenocarcinoma (PAAD). The objective was to identify a new lncRNA-miRNA-mRNA network as prognostic markers, and develop and validate a multi-mRNAs-based classifier for predicting overall survival (OS) in PAAD.

Methods

Data on pancreatic RNA expression and clinical information of 445 PAAD patients and 328 normal subjects were downloaded from The Cancer Genome Atlas (TCGA), International Cancer Genome Consortium (ICGC) and Genotype-Tissue Expression (GTEx). The weighted correlation network analysis (WGCNA) was used to analyze long non-coding RNA (lncRNA) and mRNA, clustering genes with similar expression patterns. MiRcode was used to predict the sponge microRNAs (miRNAs) corresponding to lncRNAs. The downstream targeted mRNAs of miRNAs were identified by starBase, miRDB, miRTarBase and Targetscan. A multi-mRNAs-based classifier was develop using least absolute shrinkage and selection operator method (LASSO) COX regression model, which was tested in an independent validation cohort.

Results

A lncRNA-miRNA-mRNA co-expression network which consisted of 60 lncRNAs, 3 miRNAs and 3 mRNAs associated with the prognosis of patients with PAAD was established. In addition, we constructed a 14-mRNAs-based classifier based on a training cohort composed of 178 PAAD patients, of which the area under receiver operating characteristic (AUC) in predicting 1-year, 3-year, and 5-year OS was 0.719, 0.806 and 0.794, respectively. The classifier also shown good prediction function in independent verification cohorts, with the AUC of 0.604, 0.639 and 0.607, respectively.

Conclusions

A novel competitive endogenous RNA (ceRNA) network associated with progression of PAAD could be used as a reference for future molecular biology research.