Diffusion Weighted Magnetic Resonance Imaging for the Characterization of Solitary Pulmonary Lesions

A Prospective Observational Study of Diagnostic Reliability of Semiquantitative and Quantitative High b-Value Diffusion-Weighted MRI in Distinguishing between Benign and Malignant Lung Lesions at 3 Tesla

Aim  The aim of this study was to evaluate the usefulness of high b-value diffusion-weighted imaging (DWI) to differentiate benign and malignant lung lesions in 3 Tesla magnetic resonance imaging (MRI). Materials and Methods  Thirty-one patients with lung lesions underwent a high b-value (b= 1000 s/mm 2 ) DW MRI in 3 Tesla. Thirty lesions were biopsied, followed by histopathological analysis, and one was serially followed up for 2 years. Statistical analysis was done to calculate the sensitivity, specificity, and accuracy of different DWI parameters in distinguishing benign and malignant lesions. Receiver operating characteristic (ROC) curves were used to determine the cutoff values of different parameters. Results  The qualitative assessment of signal intensity on DWI based on a 5-point rank scale had a mean score of 2.71 ± 0.75 for benign and 3. 75 ± 0.60 for malignant lesions. With a cutoff of 3.5, the sensitivity, specificity, and accuracy were 75, 86, and 77.6%, respectively. The mean ADC min (minimum apparent diffusion coefficient) value of benign and malignant lesions was 1. 49 ± 0.38 × 10-3 mm 2 /s and 1.11 ± 0.20 ×10-3 mm 2 /s, respectively. ROC curve analysis showed a cutoff value of 1.03 × 10-3 mm 2 /s; the sensitivity, specificity, and accuracy were 87.5, 71.4, and 83.3%, respectively. For lesion to spinal cord ratio and lesion to spinal cord ADC ratio with a cutoff value of 1.08 and 1.38, the sensitivity, specificity, and accuracy were 83.3 and 87.5%, 71.4 and 71.4%, and 80.6 and 83.8%, respectively. The exponential ADC showed a low accuracy rate. Conclusion  The semiquantitative and quantitative parameters of high b-value DW 3 Tesla MRI can differentiate benign from malignant lesions with high accuracy and make it a reliable nonionizing modality for characterizing lung lesions.

Predicting Non-Small-Cell Lung Cancer Survival after Curative Surgery via Deep Learning of Diffusion MRI

BACKGROUND

the objective of this study is to evaluate the predictive power of the survival model using deep learning of diffusion-weighted images (DWI) in patients with non-small-cell lung cancer (NSCLC).

METHODS

DWI at b-values of 0, 100, and 700 sec/mm2 (DWI0, DWI100, DWI700) were preoperatively obtained for 100 NSCLC patients who underwent curative surgery (57 men, 43 women; mean age, 62 years). The ADC0-100 (perfusion-sensitive ADC), ADC100-700 (perfusion-insensitive ADC), ADC0-100-700, and demographic features were collected as input data and 5-year survival was collected as output data. Our survival model adopted transfer learning from a pre-trained VGG-16 network, whereby the softmax layer was replaced with the binary classification layer for the prediction of 5-year survival. Three channels of input data were selected in combination out of DWIs and ADC images and their accuracies and AUCs were compared for the best performance during 10-fold cross validation.

RESULTS

66 patients survived, and 34 patients died. The predictive performance was the best in the following combination: DWI0-ADC0-100-ADC0-100-700 (accuracy: 92%; AUC: 0.904). This was followed by DWI0-DWI700-ADC0-100-700, DWI0-DWI100-DWI700, and DWI0-DWI0-DWI0 (accuracy: 91%, 81%, 76%; AUC: 0.889, 0.763, 0.711, respectively). Survival prediction models trained with ADC performed significantly better than the one trained with DWI only (p-values < 0.05). The survival prediction was improved when demographic features were added to the model with only DWIs, but the benefit of clinical information was not prominent when added to the best performing model using both DWI and ADC.

CONCLUSIONS

Deep learning may play a role in the survival prediction of lung cancer. The performance of learning can be enhanced by inputting precedented, proven functional parameters of the ADC instead of the original data of DWIs only.

The Efficiency of Diffusion-weighted Magnetic Resonance Imaging in the Differentiation of Malign and Benign Cavitary Lung Lesions

PURPOSE

The present study investigates the diagnostic efficiency of apparent diffusion coefficient (ADC) values in differentiating between malignant and benign cavitary lesions on diffusion-weighted magnetic resonance imaging (DWI).

MATERIALS AND METHODS

This prospective study included 45 consecutive patients identified with a cavitary lung lesion with a wall thickness of ≥5 mm on thoracic computed tomography in our clinic between 2020 and 2022, and who underwent thoracic DWI within 1 week of their original computed tomography. ADC measurements were made on DWI by drawing a region of interest manually from the cavity wall, away from the lung parenchyma in the axial section where the lesion was best demonstrated. The patients were then classified into benign and malignant groups based on the pathology or clinico-radiologic follow-up.

RESULTS

The sample included 29 (64.4%) male and 16 (35.6%) female patients, with a mean age of 59.06±17.3 years. Included in the study were 1 patient with 3 and 3 patients with 2 cavitary lesions each, with a total for the sample of 50 cavitary lesions. There were 23 (46%) malignant and 27 (54%) benign cavitary lung lesions. The mean ADC value (×10 -3  mm 2 /s) of the malignant and benign cavitary lesions was 0.977±0.522 (0.511 to 2.872) and 1.383±0.370 (0.930 to 2.213), respectively. The findings were statistically significant using an independent samples t test ( P =0.002). The mean wall thickness of the malignant and benign lesions was 12.47±5.51 mm (5 to 25 mm) and 10.11±4.65 mm (5 to 22 mm), respectively. Although malignant cavities had a higher mean wall thickness than benign cavities, the difference was statistically insignificant ( P =0.104).

CONCLUSION

A significant difference was identified between the ADC values measured in DWI of the malignant and benign cavitary lung lesions. DWI, a noninvasive and rapid imaging method, can provide useful information for the differential diagnosis of cavitary lesions and can minimize unnecessary biopsies.

Whole-lesion histogram analysis of multiple diffusion metrics for differentiating lung cancer from inflammatory lesions

Background

Whole-lesion histogram analysis can provide comprehensive assessment of tissues by calculating additional quantitative metrics such as skewness and kurtosis; however, few studies have evaluated its value in the differential diagnosis of lung lesions.

Purpose

To compare the diagnostic performance of conventional diffusion-weighted imaging (DWI), intravoxel incoherent motion (IVIM) magnetic resonance imaging (MRI) and diffusion kurtosis imaging (DKI) in differentiating lung cancer from focal inflammatory lesions, based on whole-lesion volume histogram analysis.

Methods

Fifty-nine patients with solitary pulmonary lesions underwent multiple b-values DWIs, which were then postprocessed using mono-exponential, bi-exponential and DKI models. Histogram parameters of the apparent diffusion coefficient (ADC), true diffusivity (D), pseudo-diffusion coefficient (D*), and perfusion fraction (f), apparent diffusional kurtosis (K_app) and kurtosis-corrected diffusion coefficient (D_app) were calculated and compared between the lung cancer and inflammatory lesion groups. Receiver operating characteristic (ROC) curves were constructed to evaluate the diagnostic performance.

Results

The ADC^mean, ADC^median, D ^mean and D ^median values of lung cancer were significantly lower than those of inflammatory lesions, while the ADC^skewness, K_app ^mean, K_app ^median, K_app ^SD, K_app ^kurtosis and D_app ^skewness values of lung cancer were significantly higher than those of inflammatory lesions (all p < 0.05). ADC^skewness (p = 0.019) and D ^median (p = 0.031) were identified as independent predictors of lung cancer. D ^median showed the best performance for differentiating lung cancer from inflammatory lesions, with an area under the ROC curve of 0.777. Using a D ^median of 1.091 × 10^-3 mm²/s as the optimal cut-off value, the sensitivity, specificity, positive predictive value and negative predictive value were 69.23%, 85.00%, 90.00% and 58.62%, respectively.

Conclusions

Whole-lesion histogram analysis of DWI, IVIM and DKI parameters is a promising approach for differentiating lung cancer from inflammatory lesions, and D ^median shows the best performance in the differential diagnosis of solitary pulmonary lesions.

Comparison of Conventional DWI, Intravoxel Incoherent Motion Imaging, and Diffusion Kurtosis Imaging in Differentiating Lung Lesions

Purpose

To compare conventional diffusion weighted imaging (DWI), intravoxel incoherent motion imaging (IVIM) and diffusion kurtosis imaging (DKI) in differentiating malignant and benign lung lesions.

Method

Fifty-five consecutive patients with lung lesions underwent multiple b-value DWI. The apparent diffusion coefficient (ADC), IVIM and DKI parameters were calculated using postprocessing software and compared between the malignant and benign groups. Receiver operating characteristic (ROC) analysis was performed for all parameters.

Results

ADC and D were lower in malignant lesions than in benign lesions, while Kapp was higher (P < 0.05). The differences in D*, f, and Dapp between the two groups were not significant (P > 0.05). The areas under the curves (AUCs) of ADC, D, and Kapp were 0.816, 0.864, and 0.822. The combination of all the significant parameters yielded an AUC of 0.880. There were no significant differences in diagnostic efficacy among ADC, D, Kapp and the predictor factor (PRE).

Conclusions

In this study, traditional DWI (ADC), IVIM (D), and DKI (Kapp) all had good diagnostic performance in differentiating malignant lung lesions from benign lesions, but the combination of ADC, D, and Kapp value had better diagnostic efficacy than these parameters alone.

Evaluation of DWI and ADC Sequences' Diagnostic Values in Benign and Malignant Pulmonary Lesions

OBJECTIVE

The gold standard for the diagnosis of lung cancer is conducting a histopathologic study. It is also diagnosed based on some features of a computed tomography (CT) scan. Imposed radiation is a prominent side effect of a CT scan. Diffusion-weighted imaging (DWI) apparent diffusion coefficient (ADC) images have currently been used in the diagnosis of different lesions, including those of the brain and breast, and their uses in lung lesions are being evaluated. In this study, to find a safe, sensitive, and specific method, we aimed to assess DWI imaging to replace the CT scan and the positron emission tomography scan.

MATERIAL AND METHODS

A total of 29 patients were enrolled in the study. In b800 images in DWI, spinal cord and lesion signals were measured, and the lesion-to-cord-signal ratio (LCR) was calculated. The ADC value was measured in a quantitative way. Lesions were also graded qualitatively in b800 DWI sequences.

RESULTS

There was a significant difference between malignant and benign lesions in terms of DWI grading in b800 images (p<0.001). There was a significant difference between ADC means of a malignant and benign lesion (p=0.003). The mean LCR for malignant lung lesions was significantly higher than that of the benign ones (p<0.001). Considering Grade 3 as the cutoff in DWI grading results in sensitivity, specificity, and accuracy of 89%, 90%, and 89.6%, respectively. For ADC values, sensitivity, specificity, and accuracy of 79%, 80%, and 79.3%, respectively, were obtained when the cutoff was 1.027×10^-3 sec/mm². The sensitivity of 84%, the specificity of 90%, and the accuracy of 86.2% were calculated for the LCR in a cutoff of 0.983. In this study, all three parameters had an area under the curve of ≥0.8, meaning that these variables were valuable for the differentiation of benign and malignant lesions.

CONCLUSION

Diffusion-weighted magnetic resonance imaging is a noninvasive tool, with no contrast agent and requiring ionizing radiations, which could be used for the qualitative, quantitative, and semiquantitative assessment of pulmonary lesions.

Motion correction and noise removing in lung diffusion-weighted MRI using low-rank decomposition

Lung diffusion-weighted magnetic resonance imaging (DWI) has shown a promising value in lung lesion detection, diagnosis, differentiation, and staging. However, the respiratory and cardiac motion, blood flow, and lung hysteresis may contribute to the blurring, resulting in unclear lung images. The image blurring could adversely affect diagnosis performance. The purpose of this study is to reduce the DWI blurring and assess its positive effect on diagnosis. The retrospective study includes 71 patients. In this paper, a motion correction and noise removal method using low-rank decomposition is proposed, which can reduce the DWI blurring by exploit the spatiotemporal continuity sequences. The deblurring performances are evaluated by qualitative and quantitative assessment, and the performance of diagnosis of lung cancer is measured by area under curve (AUC). In the view of the qualitative assessment, the deformation of the lung mass is reduced, and the blurring of the lung tumor edge is alleviated. Noise in the apparent diffusion coefficient (ADC) map is greatly reduced. For quantitative assessment, mutual information (MI) and Pearson correlation coefficient (Pearson-Coff) are 1.30 and 0.82 before the decomposition and 1.40 and 0.85 after the decomposition. Both the difference in MI and Pearson-Coff are statistically significant (p < 0.05). For the positive effect of deblurring on diagnosis of lung cancer, the AUC was improved from 0.731 to 0.841 using three-fold cross validation. We conclude that the low-rank matrix decomposition method is promising in reducing the errors in DWI lung images caused by noise and artifacts and improving diagnostics. Further investigations are warranted to understand the full utilities of the low-rank decomposition on lung DWI images. Graphical abstract.

Diffusion-Weighted Imaging in Oncology: An Update

To date, diffusion weighted imaging (DWI) is included in routine magnetic resonance imaging (MRI) protocols for several cancers. The real additive role of DWI lies in the "functional" information obtained by probing the free diffusivity of water molecules into intra and inter-cellular spaces that in tumors mainly depend on cellularity. Although DWI has not gained much space in some oncologic scenarios, this non-invasive tool is routinely used in clinical practice and still remains a hot research topic: it has been tested in almost all cancers to differentiate malignant from benign lesions, to distinguish different malignant histotypes or tumor grades, to predict and/or assess treatment responses, and to identify residual or recurrent tumors in follow-up examinations. In this review, we provide an up-to-date overview on the application of DWI in oncology.

Role of Various DW MRI and DCE MRI Parameters as Predictors of Malignancy in Solid Pulmonary Lesions

PURPOSE

We aimed to evaluate various diffusion and dynamic contrast-enhanced magnetic resonance imaging (DCE MRI) parameters in differentiating malignant from benign pulmonary lesions.

METHODS

We enrolled 31 (22 males) patients who had solid pulmonary lesion(s) >2 cm in our cross sectional study. Of these, 23 (74.2%) were found to be malignant on histopathology. Dynamic contrast-enhanced MRI was performed using 36 dynamic measurements (volumetric interpolated breath-hold examination). Diffusion-weighted MRI (DW MRI) performed at b value of 800 s/mm². We measured different diffusion and perfusion parameters, for example, diffusion-weighted imaging (DWI) SI, mean apparent diffusion coefficient (ADC), minimum ADC, lesion-to-spinal cord ratio, DWI score, T2 score, K^trans, K_ep, and V_e. We stratified values of each parameter as high if it was >median of values observed in our data set and low if it was ≤median. Normally distributed data were compared by unpaired t test, whereas non-normal continuous data were compared by Kruskal Wallis-H test. We applied Wilson score method to calculate sensitivity, specificity, and predictive values of parameters that were statistically significant by type of lesion with reference to histopathological examination as gold standard.

RESULTS

Diffusion-weighted imaging SI, mean ADC, minimum ADC, DWI score and K^trans values were found to be significantly different (P value < .05) by type of lesion. K^trans was found to have the highest diagnostic accuracy (74.2%) among these parameters.

CONCLUSION

K^trans and mean ADC had similar sensitivity of 65.2%. However, K^trans had highest diagnostic accuracy among various DWI and DCE MRI parameters in predicting malignancy in solid pulmonary lesions. In our study, we found a cutoff value 0.251 min^-1 for K^trans as 100% specific.

The ability and utility of diffusion-weighted magnetic resonance imaging with different 'b' values in the differentiation of benign from malignant lung lesions

Purpose

To evaluate the ability and the utility of diffusion-weighted magnetic resonance imaging (MRI) with different ‘b’ values to visualise benign and malignant lung lesions, and to determine which ‘b’ value (b = 300, 500, or 1000 s/mm²) was most useful in differentiating benign from malignant lung lesions.

Material and methods

A total of 100 patients (28 women, 72 men; mean age = 57.19 ± 13.44 years; age range = 20-83 years). Diffusion-weighted imaging (DWI) was obtained with ‘b’ values of 300, 500, and 1000 s/mm². The signal intensity of lesions on DWI images was analysed, and the apparent diffusion coefficient (ADC) values of the lesions were calculated. MRI was performed in all patients after having presented at our department for thoracic computed tomography for various reasons.

Results

A statistically significant difference in DWI signal scores was detected between benign and malignant lesions for all ‘b’ factors (p < 0.0001 for each). The sensitivity and specificity were 95% and 64%, respectively, when a score of 3 for β = 300 s/mm²; 90% and 69%, respectively, when a score of 3 for β = 500 s/mm²; and 84% and 74%, respectively, when a score of 3 for β = 1000 s/mm². ADC values showed significant differences between benign and malignant lesions for all ‘b’ factors (p < 0.0001 for each).

Conclusions

Using ‘b’ values of 300, 500, and 1000 s/mm², DWI signal intensity scores and ADC values are effective methods for the differential diagnosis of malignant and benign pulmonary lesions.

MRI in Evaluation of Solitary Pulmonary Nodules

OBJECTIVES

The aim of this study is to assess magnetic resonance imaging (MRI), diffusion-weighted imaging (DWI), T2-weighted image (T2WI), and apparent diffusion coefficient (ADC) maps' threshold values before computed tomography (CT)-guided transthorasic biopsy in solitary pulmonary nodules (SPN) by describing tumoral cell density.

MATERIALS AND METHODS

Patients who had SPN were prospectively evaluated with MRI (T1WI, T2WI) and DWI (b=0, b=500, b=1000).The ADC maps were created for each patient. Before the biopsy, lesion muscle ratios (LMR) at T2WI, ADC value, and lesion spinal cord ratio at each b values were noted. The measurements were correlated with the histopathological results.

RESULTS

A total of 53 patients were included in the study: 30.2% (n=16) were female, and 69.8% (n=37) were male. Among them, 17 lesions (32.1%) were benign, and 36 lesions (67.9%) were malignant. The age varied between 40 and 82 years, with a mean of 61.7±9.1 years. The SPN diameters were between 10 and 30 mm, and the median was 24 mm. The LSR0 and LMR values were not statistically significant in detecting malignancy. LSR500 >0.53 value can predict malignancy with 100% sensitivity and 70.6% specificity. LSR1000 >0.53 can predict malignancy with 88.9% sensitivity and 88.2% specificity. Setting the cut-off value at 0.9×10^-3, the ADC values had a sensitivity of 72.2% and a specificity of 88.2% for predicting malignancy.

CONCLUSION

For SPN follow-up, a new following-up protocol can be safely established using DWI and ADC mapping. Using these MRI parameters might decrease unnecessary biopsy rates and complications of biopsies.

Fluorine 18-FDG PET/CT and Diffusion-weighted MRI for Malignant versus Benign Pulmonary Lesions: A Meta-Analysis

Purpose To perform a meta-analysis of the literature to compare the diagnostic performance of fluorine 18 fluorodeoxyglucose PET/CT and diffusion-weighted (DW) MRI in the differentiation of malignant and benign pulmonary nodules and masses. Materials and Methods Published English-language studies on the diagnostic accuracy of PET/CT and/or DW MRI in the characterization of pulmonary lesions were searched in relevant databases through December 2017. The primary focus was on studies in which joint DW MRI and PET/CT were performed in the entire study population, to reduce interstudy heterogeneity. For DW MRI, lesion-to-spinal cord signal intensity ratio and apparent diffusion coefficient were evaluated; for PET/CT, maximum standard uptake value was evaluated. The pooled sensitivities, specificities, diagnostic odds ratios, and areas under the receiver operating characteristic curve (AUCs) for PET/CT and DW MRI were determined along with 95% confidence intervals (CIs). Results Thirty-seven studies met the inclusion criteria, with a total of 4224 participants and 4463 lesions (3090 malignant lesions [69.2%]). In the primary analysis of joint DW MRI and PET/CT studies (n = 6), DW MRI had a pooled sensitivity and specificity of 83% (95% CI: 75%, 89%) and 91% (95% CI: 80%, 96%), respectively, compared with 78% (95% CI: 70%, 84%) (P = .01 vs DW MRI) and 81% (95% CI: 72%, 88%) (P = .056 vs DW MRI) for PET/CT. DW MRI yielded an AUC of 0.93 (95% CI: 0.90, 0.95), versus 0.86 (95% CI: 0.83, 0.89) for PET/CT (P = .001). The diagnostic odds ratio of DW MRI (50 [95% CI: 19, 132]) was superior to that of PET/CT (15 [95% CI: 7, 32]) (P = .006). Conclusion The diagnostic performance of diffusion-weighted MRI is comparable or superior to that of fluorine 18 fluorodeoxyglucose PET/CT in the differentiation of malignant and benign pulmonary lesions. © RSNA, 2018 Online supplemental material is available for this article. See also the editorial by Schiebler in this issue.

Optimizing diffusion-weighted magnetic resonance imaging for evaluation of lung tumors: A comparison of respiratory triggered and free breathing techniques

Purpose

The aim of this study was to compare respiratory-triggered (RT) and free breathing (FB) diffusion weighted imaging (DWI) techniques regarding apparent diffusion coefficient (ADC) measurements and repeatability in non-squamous non-small cell lung cancer (NSCLC) measuring the total tumor volume.

Material and Methods

A total of 57 magnetic resonance imaging (MRI) examinations were analyzed. DWI was obtained by a single-shot spin-echo echo-planar imaging sequence, and for each MRI examination 2 consecutive RT and 2 consecutive FB DWI sequences were performed. Two radiologists independently read the images and made measurements. For each tumor the mean ADC value of the whole tumor volume was calculated. The difference in mean ADCs between FB and RT DWI was evaluated using the paired-sample t-test. The repeatability of ADC measurements related to imaging method was evaluated by intra class correlations (ICC) for each of the FB and RT DWI pairs.

Results

There were no significant differences in mean ADCs between FB and RT (Reader 1 p = 0.346, Reader 2 p = 0.583). The overall repeatability of ADC measurement was good for both acquisition methods, with ICCs > 0.9. Subgroup analysis showed somewhat poorer repeatability in small tumors (50 ml or less) and tumors in the lower lung zones for the RT acquisition, with ICC as low as 0.72.

Conclusions

No difference in ADC measurement or repeatability between FB and RT DWI in whole lesion ADC measurements of adenocarcinomas in the lung was demonstrated. The results imply that in this setting the FB acquisition method is accurate and possibly more robust than the RT acquisition technique.

Comparison of Various Parameters of DWI in Distinguishing Solitary Pulmonary Nodules

In order to prospectively assess various parameters of diffusion weighted imaging (DWI) in differential diagnosis of benign and malignant solitary pulmonary nodules (SPNs), 58 patients (40 men and 18 women, and mean age of 48.1±10.4 years old) with SPNs undergoing conventional MR, DWI using b=500 s/mm² on a 1.5T MR scanner, were studied. Various DWI parameters [apparent diffusion coefficient (ADC), lesion-tospinal cord signal intensity ratio (LSR), signal intensity (SI) score] were calculated and compared between malignant and benign SPNs groups. A receiver operating characteristic (ROC) curve analysis was employed to compare the diagnostic capabilities of all the parameters for discrimination between benign and malignant SPNs. The results showed that there were 42 malignant and 16 benign SPNs. The ADC was significantly lower in malignant SPNs (1.40±0.44)×10^-3 mm²/s than in benign SPNs (1.81±0.58)×10^-3 mm²/s. The LSR and SI scores were significantly increased in malignant SPNs (0.90±0.37 and 2.8±1.2) as compared with those in benign SPNs (0.68±0.39 and 2.2±1.2). The area under the ROC curves (AUC) of all parameters was not significantly different between malignant SPNs and benign SPNs. It was suggested that as three reported parameters for DWI, ADC, LSR and SI scores are all feasible for discrimination of malignant and benign SPNs. The three parameters have equal diagnostic performance.

Pulmonary MRI at 3T: Non-enhanced pulmonary magnetic resonance Imaging Characterization Quotients for differentiation of infectious and malignant lesions

OBJECTIVE

To investigate 3T pulmonary magnetic resonance imaging (MRI) for characterization of solid pulmonary lesions in immunocompromised patients and to differentiate infectious from malignant lesions.

MATERIALS AND METHODS

Thirty-eight pulmonary lesions in 29 patients were evaluated. Seventeen patients were immunocompromised (11 infections and 6 lymphomas) and 12 served as controls (4 bacterial pneumonias, 8 solid tumors). Ten of the 15 infections were acute. Signal intensities (SI) were measured in the lesion, chest wall muscle, and subcutaneous fat. Scaled SIs as Non-enhanced Imaging Characterization Quotients ((SI_Lesion-SI_Muscle)/(SI_Fat-SI_Muscle)*100) were calculated from the T2-weighted images using the mean SI (T2-NICQ_mean) or the 90th percentile of SI (T2-NICQ_90th) of the lesion. Simple quotients were calculated by dividing the SI of the lesion by the SI of chest wall muscle (e.g. T1-Q_mean: SI_Lesion/SI_Muscle).

RESULTS

Infectious pulmonary lesions showed a higher T2-NICQ_mean (40.1 [14.6-56.0] vs. 20.9 [2.4-30.1], p<0.05) and T2-NICQ_90th (74.3 [43.8-91.6] vs. 38.5 [15.8-48.1], p<0.01) than malignant lesions. T1-Q_mean was higher in malignant lesions (0.85 [0.68-0.94] vs. 0.93 [0.87-1.09], p<0.05). Considering infections only, T2-NICQ_90th was lower when anti-infectious treatment was administered >24h prior to MRI (81.8 [71.8-97.6] vs. 41.4 [26.6-51.1], p<0.01). Using Youden's index (YI), the optimal cutoff to differentiate infectious from malignant lesions was 43.1 for T2-NICQ_mean (YI=0.42, 0.47 sensitivity, 0.95 specificity) and 55.5 for T2-NICQ_90th (YI=0.61, 0.71 sensitivity, 0.91 specificity). Combining T2-NICQ_90th and T1-Q_mean increased diagnostic performance (YI=0.72, 0.77 sensitivity, 0.95 specificity).

CONCLUSION

Considering each quotient alone, T2-NICQ_90th showed the best diagnostic performance and could allow differentiation of acute infectious from malignant pulmonary lesions with high specificity. Combining T2-NICQ_90th with T1-Q_mean increased overall performance, especially regarding sensitivity.