Combination of free-breathing radial 3D fat-suppressed T1-weighted gradient-echo sequence with diffusion weighted images: Potential for differentiating malignant from benign peripheral solid pulmonary masses

Application of deep learning-based super-resolution to T1-weighted postcontrast gradient echo imaging of the chest

OBJECTIVES

A deep learning-based super-resolution for postcontrast volume-interpolated breath-hold examination (VIBE) of the chest was investigated in this study. Aim was to improve image quality, noise, artifacts and diagnostic confidence without change of acquisition parameters.

MATERIALS AND METHODS

Fifty patients who received VIBE postcontrast imaging of the chest at 1.5 T were included in this retrospective study. After acquisition of the standard VIBE (VIBE_S), a novel deep learning-based algorithm and a denoising algorithm were applied, resulting in enhanced images (VIBE_DL). Two radiologists qualitatively evaluated both datasets independently, rating sharpness of soft tissue, vessels, bronchial structures, lymph nodes, artifacts, cardiac motion artifacts, noise levels and overall diagnostic confidence, using a Likert scale ranging from 1 to 4. In the presence of lung lesions, the largest lesion was rated regarding sharpness and diagnostic confidence using the same Likert scale as mentioned above. Additionally, the largest diameter of the lesion was measured.

RESULTS

The sharpness of soft tissue, vessels, bronchial structures and lymph nodes as well as the diagnostic confidence, the extent of artifacts, the extent of cardiac motion artifacts and noise levels were rated superior in VIBE_DL (all P < 0.001). There was no significant difference in the diameter or the localization of the largest lung lesion in VIBE_DL compared to VIBE_S. Lesion sharpness as well as detectability was rated significantly better by both readers with VIBE_DL (both P < 0.001).

CONCLUSION

The application of a novel deep learning-based super-resolution approach in T1-weighted VIBE postcontrast imaging resulted in an improvement in image quality, noise levels and diagnostic confidence as well as in a shortened acquisition time.

MRI-based radiomics analysis in differentiating solid non-small-cell from small-cell lung carcinoma: a pilot study

AIM

To investigate the ability of a T2-weighted (W) magnetic resonance imaging (MRI)-based radiomics signature to differentiate solid non-small-cell lung carcinoma (NSCLC) from small-cell lung carcinoma (SCLC).

MATERIALS AND METHODS

The present retrospective study enrolled 152 eligible patients (NSCLC = 125, SCLC = 27). All patients underwent MRI using a 3 T scanner and radiomics features were extracted from T2W MRI. The least absolute shrinkage and selection operator (LASSO) logistic regression model was used to identify the optimal radiomics features for the construction of a radiomics model to differentiate solid NSCLC from SCLC. Threefold cross validation repeated 10 times was used for model training and evaluation. The conventional MRI morphology features of the lesions were also evaluated. The performance of the conventional MRI morphological features, and the radiomics signature model and nomogram model (combining radiomics signature with conventional MRI morphological features) was evaluated using receiver operating characteristic (ROC) curve analysis.

RESULTS

Five optimal features were chosen to build a radiomics signature. There was no significant difference in age, gender, and the largest diameter. The radiomics signature and conventional MRI morphological features (only pleural indentation and lymph node enlargement) were independent predictive factors for differentiating solid NSCLC from SCLC. The area under the ROC curves (AUCs) for MRI morphological features, and the radiomics model, and nomogram model was 0.69, 0.85, and 0.90 (ROC), respectively.

CONCLUSIONS

The T2W MRI-based radiomics signature is a potential non-invasive approach for distinguishing solid NSCLC from SCLC.

A pilot study of native T1-mapping for focal pulmonary lesions in 3.0 T magnetic resonance imaging: size estimation and differential diagnosis

Background

To investigate the accuracy of size estimation and potential diagnosis efficacy of native T1-mapping in focal pulmonary lesion, compared to T1-star 3D-volumetric interpolated breath-hold sequence (VIBE), T2-fBLADE turbo-spin echo (TSE), and computed tomography (CT).

Methods

Thirty-nine patients with CT-detected focal pulmonary lesions underwent thoracic 3.0-T magnetic resonance imaging (MRI) using axial free-breathing 3D T1-star VIBE, respiratory triggered T2-fBLADE TSE, breath-hold T1-Turbo fast low angle shot (FLASH) and T1-FLASH 3D. Native T1-mapping images were generated by T1-FLASH 3D with B1-filed correction by T1-Turbo FLASH. The intraclass correlation coefficient (ICC) and Bland-Altman plots were used to evaluate intra-observer agreement and inter-method reliability of diameter measurements. Native T1-values were measured and compared among the malignancy, tuberculosis, non-tuberculosis benign groups using Mann-Whitney U tests.

Results

Forty-five focal pulmonary lesions were displayed by CT, native T1-mapping, T1-star VIBE, and T2-fBLADE TSE. T1-mapping-based diameter measurements yielded an intra-observer ICC of 0.995. Additionally, inter-method measurements were highly consistent (T1-mapping & T1-star VIBE: ICC 0.982, T1-mapping & T2-fBLADE TSE: ICC 0.978, T1-mapping & CT: ICC 0.972). For lesions <3.00 cm, T1-mapping intra-observer (ICC 0.982) and inter-method diameter measurements were also highly consistent (T1-mapping & CT: ICC 0.823). Native T1-values of malignant tumors were lower than those of the non-tuberculosis benign lesions (P=0.003). Native T1-values of tuberculosis were lower than those of the non-tuberculosis benign lesions (P=0.002). Native T1-values showed no statistically significant differences between malignant tumors and tuberculosis (P=0.059).

Conclusions

Native T1-mapping enable accurate and reliable diameter measurement. Native T1-values potentially differentiate malignant tumors or tuberculosis from non-tuberculosis benign lesions.

Feasibility of free-breathing T1-weighted 3D radial VIBE for fetal MRI in various anomalies

RATIONALE AND OBJECTIVES

In magnetic resonance (MR) fetal imaging, the image quality acquired by the traditional Cartesian-sampled breath-hold T1-weighted (T1W) sequence may be degraded by motion artifacts arising from both mother and fetus. The radial VIBE sequence is reported to be a viable alternative to conventional Cartesian acquisition for both pediatric and adult MR, yielding better image quality. This study evaluated the role of radial VIBE in fetal MR imaging and compared its image quality and motion artifacts with those of the Cartesian T1W sequence.

MATERIALS AND METHODS

We included 246 pregnant women with 50 lesions on 1.5-T MR imaging. Image quality and lesion conspicuity were evaluated by two radiologists, blinded to the acquisition schemes used, using a five-point scale, where a higher score indicated a better trajectory method. Mixed-model analysis of variance and interobserver variability assessment were performed.

RESULTS

The radial VIBE sequence showed a significantly better performance than conventional T1W imaging in the head and neck, fetal body, and placenta region: 3.92 ± 0.88 vs 3 ± 0.74, p < 0.001, 3.8 ± 0.94 vs 3.15 ± 0.87, p < 0.001, and 4.17 ± 0.63 vs 3.12 ± 0.72, p < 0.001, respectively. Additionally, fewer motion artifacts were observed in all regions with the radial VIBE sequence (p < 0.01). Of 50 lesions, 49 presented better lesion conspicuity on radial VIBE images than on T1W images (4.34 ± 0.91 vs 3.48 ± 1.46, p < 0.001).

CONCLUSION

For fetal imaging, the radial VIBE sequences yielded better image quality and lesion conspicuity, with fewer motion artifacts, than conventional breath-hold Cartesian-sampled T1W sequences.