Form Factors as Potential Imaging Biomarkers to Differentiate Benign vs. Malignant Lung Lesions on CT Scans

Multimodal deep learning fusion of ultrafast-DCE MRI and clinical information for breast lesion classification

BACKGROUND

Breast cancer is the most common cancer worldwide, and magnetic resonance imaging (MRI) constitutes a very sensitive technique for invasive cancer detection. When reviewing breast MRI examination, clinical radiologists rely on multimodal information, composed of imaging data but also information not present in the images such as clinical information. Most machine learning (ML) approaches are not well suited for multimodal data. However, attention-based architectures, such as Transformers, are flexible and therefore good candidates for integrating multimodal data.

PURPOSE

The aim of this study was to develop and evaluate a novel multimodal deep learning (DL) model combining ultrafast dynamic contrast-enhanced (UF-DCE) MRI images, lesion characteristics and clinical information for breast lesion classification.

MATERIALS AND METHODS

From 2019 to 2023, UF-DCE breast images and radiology reports of 240 patients were retrospectively collected from a single clinical center and annotated. Imaging data were constituted of volumes of interest (VOI) extracted around segmented lesions. Non-imaging data were constituted of both clinical (categorical) and geometrical (scalar) data. Clinical data were extracted from annotated reports and were associated to their corresponding lesions. We compared the diagnostic performances of traditional ML methods for non-imaging data, an image model based on the DL architecture, and a novel Transformer-based architecture, the Multimodal Sieve Transformer with Vision Transformer encoder (MMST-V).

RESULTS

The final dataset included 987 lesions (280 benign, 121 malignant lesions, and 586 benign lymph nodes) and 1081 reports. For classification with non-imaging data, scalar data had a greater influence on performances of lesion classification (Area under the receiver operating characteristic curve (AUROC) = 0.875 ± 0.042) than categorical data (AUROC = 0.680 ± 0.060). MMST-V achieved better performances (AUROC = 0.928 ± 0.027) than classification based on non-imaging data (AUROC = 0.900 ± 0.045), and imaging data only (AUROC = 0.863 ± 0.025).

CONCLUSION

The proposed MMST-V is an adaptative approach that can consider redundant information provided by multimodal information. It demonstrated better performances than unimodal methods. Results highlight that the combination of clinical patient data and detailed lesion information as additional clinical knowledge enhances the diagnostic performances of UF-DCE breast MRI.

Prediction of the Benign or Malignant Nature of Pulmonary Pure Ground-Glass Nodules Based on Radiomics Analysis of High-Resolution Computed Tomography Images

To evaluate the efficacy of radiomics features extracted from preoperative high-resolution computed tomography (HRCT) scans in distinguishing benign and malignant pulmonary pure ground-glass nodules (pGGNs), a retrospective study of 395 patients from 2016 to 2020 was conducted. All nodules were randomly divided into the training and validation sets in the ratio of 7:3. Radiomics features were extracted using MaZda software (version 4.6), and the least absolute shrinkage and selection operator (LASSO) was employed for feature selection. Significant differences were observed in the training set between benign and malignant pGGNs in sex, mean CT value, margin, pleural retraction, tumor-lung interface, and internal vascular change, and then the mean CT value and the morphological features model were constructed. Fourteen radiomics features were selected by LASSO for the radiomics model. The combined model was developed by integrating all selected radiographic and radiomics features using logistic regression. The AUCs in the training set were 0.606 for the mean CT value, 0.718 for morphological features, 0.756 for radiomics features, and 0.808 for the combined model. In the validation set, AUCs were 0.601, 0.692, 0.696, and 0.738, respectively. The decision curves showed that the combined model demonstrated the highest net benefit.

A diagnostic biomarker of acid glycoprotein 1 for distinguishing malignant from benign pulmonary lesions

BACKGROUND

The acid glycoprotein 1 (AGP1) is downregulated in lung cancer. However, the performance of AGP1 in distinguishing benign from malignant lung lesions is still unknown.

METHODS

The expression of AGP1 in benign diseases and lung cancer samples was detected by Western blot. The receiver operating characteristic curves, bivariate correlation, and multivariate analysis was analyzed by SPSS software.

RESULTS

AGP1 expression levels were significantly downregulated in lung cancer and correlated with carcinoembryonic antigen (CEA), CA199, and CA724 tumor biomarkers. The diagnostic performance of AGP1 for distinguishing malignant from benign pulmonary lesions was better than the other four clinical biomarkers including CEA, squamous cell carcinoma-associated antigen, neuron-specific enolase, and cytokeratin 19 fragment 21-1, with an area under the curve value of 0.713 at 88.8% sensitivity. Furthermore, the multivariate analysis indicated that the variates of thrombin time and potassium significantly affected the AGP1 levels in lung cancer.

CONCLUSIONS

Our study indicates that AGP1 expression is decreased in lung cancer compared to benign samples, which helps distinguish benign and malignant pulmonary lesions.

Performance Analysis of Six Semi-Automated Tumour Delineation Methods on [18F] Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography (FDG PET/CT) in Patients with Head and Neck Cancer

Background. Head and neck cancer (HNC) is the seventh most common neoplastic disorder at the global level. Contouring HNC lesions on [18F] Fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT) scans plays a fundamental role for diagnosis, risk assessment, radiotherapy planning and post-treatment evaluation. However, manual contouring is a lengthy and tedious procedure which requires significant effort from the clinician. Methods. We evaluated the performance of six hand-crafted, training-free methods (four threshold-based, two algorithm-based) for the semi-automated delineation of HNC lesions on FDG PET/CT. This study was carried out on a single-centre population of n=103 subjects, and the standard of reference was manual segmentation generated by nuclear medicine specialists. Figures of merit were the Sørensen-Dice coefficient (DSC) and relative volume difference (RVD). Results. Median DSC ranged between 0.595 and 0.792, median RVD between -22.0% and 87.4%. Click and draw and Nestle's methods achieved the best segmentation accuracy (median DSC, respectively, 0.792 ± 0.178 and 0.762 ± 0.107; median RVD, respectively, -21.6% ± 1270.8% and -32.7% ± 40.0%) and outperformed the other methods by a significant margin. Nestle's method also resulted in a lower dispersion of the data, hence showing stronger inter-patient stability. The accuracy of the two best methods was in agreement with the most recent state-of-the art results. Conclusions. Semi-automated PET delineation methods show potential to assist clinicians in the segmentation of HNC lesions on FDG PET/CT images, although manual refinement may sometimes be needed to obtain clinically acceptable ROIs.

Evaluation of calcification distribution by CT-based textural analysis for discrimination of immature teratoma

BACKGROUND

Mature and immature teratomas are differentiated based on tumor markers and calcification or fat distribution. However, no study has objectively quantified the differences in calcification and fat distributions between these tumors. This study aimed to evaluate the diagnostic potential of CT-based textural analysis in differentiating between mature and immature teratomas in patients aged < 20 years.

MATERIALS AND METHODS

Thirty-two patients with pathologically proven mature cystic (n = 28) and immature teratomas (n = 4) underwent transabdominal ultrasound and/or abdominal and pelvic CT before surgery. The diagnostic performance of CT for assessing imaging features, including subjective manual measurement and objective textural analysis of fat and calcification distributions in the tumors, was evaluated by two experienced readers. The histopathological results were used as the gold standard. The Mann-Whitney U test was used for statistical analysis.

RESULTS

We evaluated 32 patients (mean age, 14.5 years; age range, 6-19 years). The mean maximum diameter and number of calcifications of immature teratomas were significantly larger than those of mature cystic teratomas (p < 0.01). The mean number of fats of immature teratomas was significantly larger than that of mature cystic teratomas (p < 0.01); however, no significant difference in the maximum diameter of fats was observed. CT textural features for calcification distribution in the tumors showed that mature cystic teratomas had higher homogeneity and energy than immature teratomas. However, immature teratomas showed higher correlation, entropy, and dissimilarity than mature cystic teratomas among features derived from the gray-level co-occurrence matrix (GLCM) (p < 0.05). No significant differences were observed in the CT features of fats derived from GLCM.

CONCLUSION

Our results demonstrate that calcification distribution on CT is a potential diagnostic biomarker to discriminate mature from immature teratomas, thus enabling optimal therapeutic selection for patients aged < 20 years.

PET Radiomics and Response to Immunotherapy in Lung Cancer: A Systematic Review of the Literature

The aim of this review is to provide a comprehensive overview of the existing literature concerning the applications of positron emission tomography (PET) radiomics in lung cancer patient candidates or those undergoing immunotherapy.

MATERIALS AND METHODS

A systematic review was conducted on databases and web sources. English-language original articles were considered. The title and abstract were independently reviewed to evaluate study inclusion. Duplicate, out-of-topic, and review papers, or editorials, articles, and letters to editors were excluded. For each study, the radiomics analysis was assessed based on the radiomics quality score (RQS 2.0). The review was registered on the PROSPERO database with the number CRD42023402302.

RESULTS

Fifteen papers were included, thirteen were qualified as using conventional radiomics approaches, and two used deep learning radiomics. The content of each study was different; indeed, seven papers investigated the potential ability of radiomics to predict PD-L1 expression and tumor microenvironment before starting immunotherapy. Moreover, two evaluated the prediction of response, and four investigated the utility of radiomics to predict the response to immunotherapy. Finally, two papers investigated the prediction of adverse events due to immunotherapy.

CONCLUSIONS

Radiomics is promising for the evaluation of TME and for the prediction of response to immunotherapy, but some limitations should be overcome.

Diagnosis of malignant pleural effusion with combinations of multiple tumor markers: A comparison study of five machine learning models

BACKGROUND

To evaluate the diagnostic value of combinations of tumor markers carcinoembryonic antigen (CEA), carbohydrate antigen (CA) 125, CA153, and CA19-9 in identifying malignant pleural effusion (MPE) from non-malignant pleural effusion (non-MPE) using machine learning, and compare the performance of popular machine learning methods.

METHODS

A total of 319 samples were collected from patients with pleural effusion in Beijing and Wuhan, China, from January 2018 to June 2020. Five machine learning methods including Logistic regression, extreme gradient boosting (XGBoost), Bayesian additive regression tree, random forest, and support vector machine were applied to evaluate the diagnostic performance. Sensitivity, specificity, Youden's index, and the area under the receiver operating characteristic curve (AUC) were used to evaluate the performance of different diagnostic models.

RESULTS

For diagnostic models with a single tumor marker, the model using CEA, constructed by XGBoost, performed best (AUC = 0.895, sensitivity = 0.80), and the model with CA153, also by XGBoost, showed the largest specificity 0.98. Among all combinations of tumor markers, the combination of CEA and CA153 achieved the best performance (AUC = 0.921, sensitivity = 0.85) in identifying MPE under the diagnostic model constructed by XGBoost.

CONCLUSIONS

Diagnostic models for MPE with a combination of multiple tumor markers outperformed the models with a single tumor marker, particularly in sensitivity. Using machine learning methods, especially XGBoost, could comprehensively improve the diagnostic accuracy of MPE.

Qualitative and Semiquantitative Parameters of 18F-FDG-PET/CT as Predictors of Malignancy in Patients with Solitary Pulmonary Nodule

This study aims to evaluate the reliability of qualitative and semiquantitative parameters of ¹⁸F-FDG PET-CT, and eventually a correlation between them, in predicting the risk of malignancy in patients with solitary pulmonary nodules (SPNs) before the diagnosis of lung cancer. A total of 146 patients were retrospectively studied according to their pre-test probability of malignancy (all patients were intermediate risk), based on radiological features and risk factors, and qualitative and semiquantitative parameters, such as SUVmax, SUVmean, TLG, and MTV, which were obtained from the FDG PET-CT scan of such patients before diagnosis. It has been observed that visual analysis correlates well with the risk of malignancy in patients with SPN; indeed, only 20% of SPNs in which FDG uptake was low or absent were found to be malignant at the cytopathological examination, while 45.45% of SPNs in which FDG uptake was moderate and 90.24% in which FDG uptake was intense were found to be malignant. The same trend was observed evaluating semiquantitative parameters, since increasing values of SUVmax, SUVmean, TLG, and MTV were observed in patients whose cytopathological examination of SPN showed the presence of lung cancer. In particular, in patients whose SPN was neoplastic, we observed a median (MAD) SUVmax of 7.89 (±2.24), median (MAD) SUVmean of 3.76 (±2.59), median (MAD) TLG of 16.36 (±15.87), and a median (MAD) MTV of 3.39 (±2.86). In contrast, in patients whose SPN was non-neoplastic, the SUVmax was 2.24 (±1.73), SUVmean 1.67 (±1.15), TLG 1.63 (±2.33), and MTV 1.20 (±1.20). Optimal cut-offs were drawn for semiquantitative parameters considered predictors of malignancy. Nodule size correlated significantly with FDG uptake intensity and with SUVmax. Finally, age and nodule size proved significant predictors of malignancy. In conclusion, considering the pre-test probability of malignancy, qualitative and semiquantitative parameters can be considered reliable tools in patients with SPN, since cut-offs for SUVmax, SUVmean, TLG, and MTV showed good sensitivity and specificity in predicting malignancy.

Diagnostic Accuracy of the Artificial Intelligence Methods in Medical Imaging for Pulmonary Tuberculosis: A Systematic Review and Meta-Analysis

Tuberculosis (TB) remains one of the leading causes of death among infectious diseases worldwide. Early screening and diagnosis of pulmonary tuberculosis (PTB) is crucial in TB control, and tend to benefit from artificial intelligence. Here, we aimed to evaluate the diagnostic efficacy of a variety of artificial intelligence methods in medical imaging for PTB. We searched MEDLINE and Embase with the OVID platform to identify trials published update to November 2022 that evaluated the effectiveness of artificial-intelligence-based software in medical imaging of patients with PTB. After data extraction, the quality of studies was assessed using quality assessment of diagnostic accuracy studies 2 (QUADAS-2). Pooled sensitivity and specificity were estimated using a bivariate random-effects model. In total, 3987 references were initially identified and 61 studies were finally included, covering a wide range of 124,959 individuals. The pooled sensitivity and the specificity were 91% (95% confidence interval (CI), 89-93%) and 65% (54-75%), respectively, in clinical trials, and 94% (89-96%) and 95% (91-97%), respectively, in model-development studies. These findings have demonstrated that artificial-intelligence-based software could serve as an accurate tool to diagnose PTB in medical imaging. However, standardized reporting guidance regarding AI-specific trials and multicenter clinical trials is urgently needed to truly transform this cutting-edge technology into clinical practice.