A non-invasive, automated diagnosis of Menière's disease using radiomics and machine learning on conventional magnetic resonance imaging: A multicentric, case-controlled feasibility study

Insights into radiomics: a comprehensive review for beginners

Radiomics and artificial intelligence (AI) are rapidly evolving, significantly transforming the field of medical imaging. Despite their growing adoption, these technologies remain challenging to approach due to their technical complexity. This review serves as a practical guide for early-career radiologists and researchers seeking to integrate radiomics into their studies. It provides practical insights for clinical and research applications, addressing common challenges, limitations, and future directions in the field. This work offers a structured overview of the essential steps in the radiomics workflow, focusing on concrete aspects of each step, including indicative and practical examples. It covers the main steps such as dataset definition, image acquisition and preprocessing, segmentation, feature extraction and selection, and AI model training and validation. Different methods to be considered are discussed, accompanied by summary diagrams. This review equips readers with the knowledge necessary to approach radiomics and AI in medical imaging from a hands-on research perspective.

Risk evaluation and incidence prediction of endolymphatic hydrops using multilayer perceptron in patients with audiovestibular symptoms

Endolymphatic hydrops (EH) has been visualized on magnetic resonance imaging (MRI) in patients with various inner ear diseases. The purpose of this study was to evaluate the prevalence and risk factors of significant EH on inner ear MRI in patients with 1 or more audiovestibular symptoms and to predict the incidence of significant EH using multivariate analysis and multilayer perceptron artificial neural network modeling. This retrospective study included a total of 135 patients with 1 or more audiovestibular symptoms who do not meet the diagnostic criteria for MD and underwent inner ear MRI at our institution from July 2021 to January 2024. The EH grade of each patient was evaluated, and "significant EH" was considered grade II or III. Of 135 patients with 1 or more audiovestibular symptoms, 48 patients (35.6%) presented with significant EH and 87 patients (64.4%) without significant EH on inner ear MRI. The prevalence of significant EH was higher in males, which was statistically significant (P = .007). The prevalence of significant EH was higher in the right ear, and the mean age of patients with significant EH was 1.94 years higher, but no statistical significance was observed (P = .660 and .456, retrospectively). The odds ratio for significant EH development was 2.696 (95% confidence interval: 1.296-5.607) times higher in men, which was statistically significant. Predicting the incidence of significant EH development using multivariate analysis, sex was the only variable that was statistically significant (P = .008). Based on a predictive model using multilayer perceptron (MLP), the classification accuracy of the model was 79.5%. In our study, the male gender could be related to the risk of developing significant EH in patients with audiovestibular symptoms. The accuracy of our suggested MLP model for predicting the incidence of significant EH was 79.5%, with sex being the highest predictor importance. In the future, inner ear MRI and MLP neural network modeling can be combined as a noninvasive and precise support system in the diagnosis of EH.

Non-Invasive Biomarkers in the Era of Big Data and Machine Learning

Invasive diagnostic techniques, while offering critical insights into disease pathophysiology, are often limited by high costs, procedural risks, and patient discomfort. Non-invasive biomarkers represent a transformative alternative, providing diagnostic precision through accessible biological samples or physiological data, including blood, saliva, breath, and wearable health metrics. They encompass molecular and imaging approaches, revealing genetic, epigenetic, and metabolic alterations associated with disease states. Furthermore, advances in breathomics and gut microbiome profiling further expand their diagnostic scope. Even with their strengths in terms of safety, cost-effectiveness, and accessibility, non-invasive biomarkers face challenges in achieving monitoring sensitivity and specificity comparable to traditional clinical approaches. Computational advancements, particularly in artificial intelligence and machine learning, are addressing these limitations by uncovering complex patterns in multi-modal datasets, enhancing diagnostic accuracy and facilitating personalized medicine. The present review integrates recent innovations, examines their clinical applications, highlights their limitations and provides a concise overview of the evolving role of non-invasive biomarkers in precision diagnostics, positioning them as a compelling choice for large-scale healthcare applications.

Applications of Machine Learning in Meniere's Disease Assessment Based on Pure-Tone Audiometry

OBJECTIVE

To apply machine learning models based on air conduction thresholds of pure-tone audiometry for automatic diagnosis of Meniere's disease (MD) and prediction of endolymphatic hydrops (EH).

STUDY DESIGN

Retrospective study.

SETTING

Tertiary medical center.

METHODS

Gadolinium-enhanced magnetic resonance imaging sequences and pure-tone audiometry data were collected. Subsequently, basic and multiple analytical features were engineered based on the air conduction thresholds of pure-tone audiometry. Later, 5 classical machine learning models were trained to diagnose MD using the engineered features. The models demonstrating excellent performance were also selected to predict EH. The model's effectiveness in MD diagnosis was compared with experienced otolaryngologists.

RESULTS

First, the winning light gradient boosting (LGB) machine learning model trained by multiple features demonstrates a remarkable performance on the diagnosis of MD, achieving an accuracy rate of 87%, sensitivity of 83%, specificity of 90%, and a robust area under the receiver operating characteristic curve of 0.95, which compares favorably with experienced clinicians. Second, the LGB model, with an accuracy of 78% on EH prediction, outperformed the other 3 machine learning models. Finally, a feature importance analysis reveals a pivotal role of the specific pure-tone audiometry features that are essential for both MD diagnosis and EH prediction. Highlighted features include standard deviation and mean of the whole-frequency hearing, the peak of the audiogram, and hearing at low frequencies, notably at 250 Hz.

CONCLUSION

An efficient machine learning model based on pure-tone audiometry features was produced to diagnose MD, which also showed the potential to predict the subtypes of EH. The innovative approach demonstrated a game-changing strategy for MD screening and promising cost-effective benefits for the health care enterprise.

Computer Vision-Radiomics & Pathognomics

The role of computer vision in extracting radiographic (radiomics) and histopathologic (pathognomics) features is an extension of molecular biomarkers that have been foundational to our understanding across the spectrum of head and neck disorders. Especially within head and neck cancers, machine learning and deep learning applications have yielded advances in the characterization of tumor features, nodal features, and various outcomes. This review aims to overview the landscape of radiomic and pathognomic applications, informing future work to address gaps. Novel methodologies will be needed to potentially engineer ways of integrating multidimensional data inputs to examine disease features to guide prognosis comprehensively and ultimately clinical management.

The relationship between endolymphatic hydrops features and hearing loss in Bilateral Meniere's disease

BACKGROUND

This study aimed to investigate the relationship between the features of endolymphatic hydrops and hearing loss in patients with Bilateral Meniere's Disease.

METHODS

A retrospective analysis was conducted on 77 patients diagnosed with Bilateral Meniere's Disease. The features of endolymphatic hydrops in the affected ear were evaluated through gadolinium-enhanced inner ear Magnetic resonance imaging. The Spearman correlation coefficient, paired t-tests, and Wilcoxon signed-rank tests were employed for data analysis.

RESULTS

The analysis revealed a significant correlation between the degree of endolymphatic hydrops and hearing loss across all frequencies(0.125-8 kHz), including the cochlear, vestibular, and overall degree of endolymphatic hydrops. The strongest correlation between the overall degree of endolymphatic hydrops and hearing loss was observed at low frequencies (r = 0.571, p < 0.05), followed by mid-frequencies (r = 0.508, p < 0.05), and high-frequencies (r = 0.351, p < 0.05), with a correlation of r = 0.463, p < 0.05 for the staging of Meniere's disease. Affected Ears with endolymphatic hydrops both in the cochlea and vestibule exhibited more severe hearing loss and Meniere's disease staging compared to those with isolated endolymphatic hydrops within the same patient.

CONCLUSIONS

The features of endolymphatic hydrops in patients with Bilateral Meniere's Disease were found to correlate with the severity of hearing loss and the staging of Meniere's disease.

A novel radiomics nomogram based on T2-sampling perfection with application-optimized contrasts using different flip-angle evolutions (SPACE) images for predicting cochlear and vestibular endolymphatic hydrops in Meniere's disease patients

OBJECTIVES

To construct and validate a radiomics nomogram based on T2-sampling perfection with application-optimized contrasts using different flip-angle evolutions (SPACE) images for predicting cochlear and vestibular endolymphatic hydrops (EH) in Meniere's disease patients.

METHODS

A total of 156 patients (312 affected ears) with bilateral definite Meniere's disease who underwent delayed enhancement MRI scans were enrolled in this study. All ears of the patients were divided into a training set (n = 218) and an internal validation set (n = 94). A radiomics nomogram was constructed from radiomics features extracted from the T2-SPACE images, and a radiomics score was calculated. Performance of the radiomics nomogram was assessed using receiver operating characteristics analysis.

RESULTS

Five features were selected for the construction of the cochlear radiomics nomogram, and seven features for the vestibular radiomics nomogram. The radiomics nomograms exhibited robust performance in differentiating between EH-positive and EH-negative statuses in both training and validation cohorts, with the area under the receiver operating characteristics curve values for cochlear and vestibular radiomic nomograms being 0.703 and 0.728 in the training set, and 0.718 and 0.701 in the validation set, respectively.

CONCLUSION

The novel radiomics nomograms based on T2-SPACE images were successfully constructed to predict cochlear and vestibular EH in Meniere's disease. The models showed a solid and superior performance and may play an important role in the EH prediction.

CLINICAL RELEVANCE STATEMENT

We constructed a novel radiomics nomogram, which can be a very useful tool for predicting cochlear and vestibular endolymphatic hydrops in Meniere's disease patients.

KEY POINTS

• This is the first T2-SPACE-based nomogram to predict cochlear and vestibular endolymphatic hydrops. • The nomogram is of great value to patients who are unable to undergo delayed enhancement MRI scans.

Noninvasive prediction of lymph node metastasis in pancreatic cancer using an ultrasound-based clinicoradiomics machine learning model

OBJECTIVES

This study was designed to explore and validate the value of different machine learning models based on ultrasound image-omics features in the preoperative diagnosis of lymph node metastasis in pancreatic cancer (PC).

METHODS

This research involved 189 individuals diagnosed with PC confirmed by surgical pathology (training cohort: n = 151; test cohort: n = 38), including 50 cases of lymph node metastasis. Image-omics features were extracted from ultrasound images. After dimensionality reduction and screening, eight machine learning algorithms, including logistic regression (LR), support vector machine (SVM), K-nearest neighbors (KNN), random forest (RF), extra trees (ET), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and multilayer perceptron (MLP), were used to establish image-omics models to predict lymph node metastasis in PC. The best omics prediction model was selected through ROC curve analysis. Machine learning models were used to analyze clinical features and determine variables to establish a clinical model. A combined model was constructed by combining ultrasound image-omics and clinical features. Decision curve analysis (DCA) and a nomogram were used to evaluate the clinical application value of the model.

RESULTS

A total of 1561 image-omics features were extracted from ultrasound images. 15 valuable image-omics features were determined by regularization, dimension reduction, and algorithm selection. In the image-omics model, the LR model showed higher prediction efficiency and robustness, with an area under the ROC curve (AUC) of 0.773 in the training set and an AUC of 0.850 in the test set. The clinical model constructed by the boundary of lesions in ultrasound images and the clinical feature CA199 (AUC = 0.875). The combined model had the best prediction performance, with an AUC of 0.872 in the training set and 0.918 in the test set. The combined model showed better clinical benefit according to DCA, and the nomogram score provided clinical prediction solutions.

CONCLUSION

The combined model established with clinical features has good diagnostic ability and can be used to predict lymph node metastasis in patients with PC. It is expected to provide an effective noninvasive method for clinical decision-making, thereby improving the diagnosis and treatment of PC.

Potential Application of Hydrops MR Imaging: A Systematic Review

BACKGROUND

Diagnostic dilemma between clinical Meniere's disease and radiological endolymphatic hydrops (EH) has emerged since the introduction of hydrops magnetic resonance imaging (MRI). The aim of this study is to explore the potential application of hydrops MRI on diagnosing the EH.

METHODS

This review was developed from peer-reviewed articles published in those journals listed on journal of citation reports. The MEDLINE database of the US National Library of Medicine, Scopus, and Google Scholar were used to collect articles based on the guidelines (PRISMA 2020 statement) for reporting reviews.

RESULTS

Initially, 470 articles were retrieved from 1983 to 2023, and 80 relevant articles were ultimately selected. The sensitivity (69%-92%) and specificity (78%-96%) values varied from each laboratory for detecting EH via hydrops MRI, probably due to candidate selection and the grading system employed.

CONCLUSION

The application of hydrops MRI allows (1) differentiation between EH and sudden sensorineural hearing loss; (2) determination of the affected side of EH; and (3) confirmation of the diagnosis of EH concomitant with other disorders. Notably, not all differentials for EH can be visualized on MR images. One of the existing gaps to be filled is that updated hydrops MRI fails to identify distortion, that is, rupture, collapse, fistula, or fibrosis of the inner ear compartments, akin to what histopathological evidence can demonstrate. Hence, enhanced ultrahigh resolution of hydrops MRI is required for demonstrating fine structures of the inner ear compartments in the future.

Prognostic Assessment of Gastropancreatic Neuroendocrine Neoplasm: Prospects and Limits of Radiomics

Neuroendocrine neoplasms (NENs) are a group of lesions originating from cells of the diffuse neuroendocrine system. NENs may involve different sites, including the gastrointestinal tract (GEP-NENs). The incidence and prevalence of GEP-NENs has been constantly rising thanks to the increased diagnostic power of imaging and immuno-histochemistry. Despite the plethora of biochemical markers and imaging techniques, the prognosis and therapeutic choice in GEP-NENs still represents a challenge, mainly due to the great heterogeneity in terms of tumor lesions and clinical behavior. The concept that biomedical images contain information about tissue heterogeneity and pathological processes invisible to the human eye is now well established. From this substrate comes the idea of radiomics. Computational analysis has achieved promising results in several oncological settings, and the use of radiomics in different types of GEP-NENs is growing in the field of research, yet with conflicting results. The aim of this narrative review is to provide a comprehensive update on the role of radiomics on GEP-NEN management, focusing on the main clinical aspects analyzed by most existing reports: predicting tumor grade, distinguishing NET from other tumors, and prognosis assessment.

Mass Spectrometric Interrogation of Earwax: Toward the Detection of Ménière's Disease

Many diseases remain difficult to identify because the occurrence of characteristic biomarkers within traditional matrices such as blood and urine remain unknown. Disease diagnosis could, therefore, benefit from the analysis of readily accessible, non-traditional matrices that have a high chemical content and contain distinguishing biomarkers. One such matrix is cerumen (i.e., earwax), whose chemical complexity can pose challenges when analyzed by conventional methods. A combination of cerumen chemical profiles analyzed by gas chromatography-mass spectrometry (GC-MS) and direct analysis in real time-high-resolution mass spectrometry (DART-HRMS) were investigated to ascertain the possible presence of the rare otolaryngological disorder Ménière's disease. This illness is currently identified via "diagnosis by exclusion" in which the disease is distinguished from others with overlapping symptoms by the process of elimination. GC-MS revealed a chemical profile difference between those with and without a Ménière's disease diagnosis by a visually apparent diminution of the compounds present in the Ménière's disease samples. DART-HRMS revealed that the two classes could be differentiated using three fatty acids: cis-9-hexadecenoic acid, cis-10-heptadecenoic acid, and cis-9-octadecenoic acid. These compounds were subsequently quantified by GC-MS and overall, the amounts of these fatty acids were decreased in Ménière's disease patients. The average levels for non-Ménière's disease samples were 7.89 μg/mg for cis-9-hexadecenoic acid, 0.87 μg/mg for cis-10-heptadecenoic acid, and 4.94 μg/mg for cis-9-octadecenoic acid. The average levels for Ménière's disease samples were 1.70 μg/mg for cis-9-hexadecenoic acid, 0.13 μg/mg for cis-10-heptadecenoic acid, and 2.07 μg/mg for cis-9-octadecenoic acid. The confidence levels for cis-9-hexadecenoic acid, cis-10-heptadecenoic acid, and cis-9-octadecenoic acid were 98.7%, 99.9%, and 95.4%, respectively. The results suggest that assessment of the concentrations of these fatty acids could be a useful clinical tool for the more rapid and accurate detection of Ménière's disease.

Peritoneal Carcinosis: What the Radiologist Needs to Know

Peritoneal carcinosis is a condition characterized by the spread of cancer cells to the peritoneum, which is the thin membrane that lines the abdominal cavity. It is a serious condition that can result from many different types of cancer, including ovarian, colon, stomach, pancreatic, and appendix cancer. The diagnosis and quantification of lesions in peritoneal carcinosis are critical in the management of patients with the condition, and imaging plays a central role in this process. Radiologists play a vital role in the multidisciplinary management of patients with peritoneal carcinosis. They need to have a thorough understanding of the pathophysiology of the condition, the underlying neoplasms, and the typical imaging findings. In addition, they need to be aware of the differential diagnoses and the advantages and disadvantages of the various imaging methods available. Imaging plays a central role in the diagnosis and quantification of lesions, and radiologists play a critical role in this process. Ultrasound, computed tomography, magnetic resonance, and PET/CT scans are used to diagnose peritoneal carcinosis. Each imaging procedure has advantages and disadvantages, and particular imaging techniques are recommended based on patient conditions. Our aim is to provide knowledge to radiologists regarding appropriate techniques, imaging findings, differential diagnoses, and treatment options. With the advent of AI in oncology, the future of precision medicine appears promising, and the interconnection between structured reporting and AI is likely to improve diagnostic accuracy and treatment outcomes for patients with peritoneal carcinosis.

Update on the Applications of Radiomics in Diagnosis, Staging, and Recurrence of Intrahepatic Cholangiocarcinoma

BACKGROUND

This paper offers an assessment of radiomics tools in the evaluation of intrahepatic cholangiocarcinoma.

METHODS

The PubMed database was searched for papers published in the English language no earlier than October 2022.

RESULTS

We found 236 studies, and 37 satisfied our research criteria. Several studies addressed multidisciplinary topics, especially diagnosis, prognosis, response to therapy, and prediction of staging (TNM) or pathomorphological patterns. In this review, we have covered diagnostic tools developed through machine learning, deep learning, and neural network for the recurrence and prediction of biological characteristics. The majority of the studies were retrospective.

CONCLUSIONS

It is possible to conclude that many performing models have been developed to make differential diagnosis easier for radiologists to predict recurrence and genomic patterns. However, all the studies were retrospective, lacking further external validation in prospective and multicentric cohorts. Furthermore, the radiomics models and the expression of results should be standardized and automatized to be applicable in clinical practice.

Machine learning at the edge for AI-enabled multiplexed pathogen detection

Multiplexed detection of biomarkers in real-time is crucial for sensitive and accurate diagnosis at the point of use. This scenario poses tremendous challenges for detection and identification of signals of varying shape and quality at the edge of the signal-to-noise limit. Here, we demonstrate a robust target identification scheme that utilizes a Deep Neural Network (DNN) for multiplex detection of single particles and molecular biomarkers. The model combines fast wavelet particle detection with Short-Time Fourier Transform analysis, followed by DNN identification on an AI-specific edge device (Google Coral Dev board). The approach is validated using multi-spot optical excitation of Klebsiella Pneumoniae bacterial nucleic acids flowing through an optofluidic waveguide chip that produces fluorescence signals of varying amplitude, duration, and quality. Amplification-free 3× multiplexing in real-time is demonstrated with excellent specificity, sensitivity, and a classification accuracy of 99.8%. These results show that a minimalistic DNN design optimized for mobile devices provides a robust framework for accurate pathogen detection using compact, low-cost diagnostic devices.

MRI-based artificial intelligence to predict infection following total hip arthroplasty failure

PURPOSE

To investigate whether artificial intelligence (AI) can differentiate septic from non-septic total hip arthroplasty (THA) failure based on preoperative MRI features.

MATERIALS AND METHODS

We included 173 patients (98 females, age: 67 ± 12 years) subjected to first-time THA revision surgery after preoperative pelvis MRI. We divided the patients into a training/validation/internal testing cohort (n = 117) and a temporally independent external-testing cohort (n = 56). MRI features were used to train, validate and test a machine learning algorithm based on support vector machine (SVM) to predict THA infection on the training-internal validation cohort with a nested fivefold validation approach. Machine learning performance was evaluated on independent data from the external-testing cohort.

RESULTS

MRI features were significantly more frequently observed in THA infection (P < 0.001), except bone destruction, periarticular soft-tissue mass, and fibrous membrane (P > 0.005). Considering all MRI features in the training/validation/internal-testing cohort, SVM classifier reached 92% sensitivity, 62% specificity, 79% PPV, 83% NPV, 82% accuracy, and 81% AUC in predicting THA infection, with bone edema, extracapsular edema, and synovitis having been the best predictors. After being tested on the external-testing cohort, the classifier showed 92% sensitivity, 79% specificity, 89% PPV, 83% NPV, 88% accuracy, and 89% AUC in predicting THA infection. SVM classifier showed 81% sensitivity, 76% specificity, 66% PPV, 88% NPV, 80% accuracy, and 74% AUC in predicting THA infection in the training/validation/internal-testing cohort based on the only presence of periprosthetic bone marrow edema on MRI, while it showed 68% sensitivity, 89% specificity, 93% PPV, 60% NPV, 75% accuracy, and 79% AUC in the external-testing cohort.

CONCLUSION

AI using SVM classifier showed promising results in predicting THA infection based on MRI features. This model might support radiologists in identifying THA infection.

A Narrative Review on LI-RADS Algorithm in Liver Tumors: Prospects and Pitfalls

Liver cancer is the sixth most detected tumor and the third leading cause of tumor death worldwide. Hepatocellular carcinoma (HCC) is the most common primary liver malignancy with specific risk factors and a targeted population. Imaging plays a major role in the management of HCC from screening to post-therapy follow-up. In order to optimize the diagnostic-therapeutic management and using a universal report, which allows more effective communication among the multidisciplinary team, several classification systems have been proposed over time, and LI-RADS is the most utilized. Currently, LI-RADS comprises four algorithms addressing screening and surveillance, diagnosis on computed tomography (CT)/magnetic resonance imaging (MRI), diagnosis on contrast-enhanced ultrasound (CEUS) and treatment response on CT/MRI. The algorithm allows guiding the radiologist through a stepwise process of assigning a category to a liver observation, recognizing both major and ancillary features. This process allows for characterizing liver lesions and assessing treatment. In this review, we highlighted both major and ancillary features that could define HCC. The distinctive dynamic vascular pattern of arterial hyperenhancement followed by washout in the portal-venous phase is the key hallmark of HCC, with a specificity value close to 100%. However, the sensitivity value of these combined criteria is inadequate. Recent evidence has proven that liver-specific contrast could be an important tool not only in increasing sensitivity but also in diagnosis as a major criterion. Although LI-RADS emerges as an essential instrument to support the management of liver tumors, still many improvements are needed to overcome the current limitations. In particular, features that may clearly distinguish HCC from cholangiocarcinoma (CCA) and combined HCC-CCA lesions and the assessment after locoregional radiation-based therapy are still fields of research.

Radiomics and machine learning analysis based on magnetic resonance imaging in the assessment of liver mucinous colorectal metastases

PURPOSE

The purpose of this study is to evaluate the Radiomics and Machine Learning Analysis based on MRI in the assessment of Liver Mucinous Colorectal Metastases.Query METHODS: The cohort of patients included a training set (121 cases) and an external validation set (30 cases) with colorectal liver metastases with pathological proof and MRI study enrolled in this approved study retrospectively. About 851 radiomics features were extracted as median values by means of the PyRadiomics tool on volume on interest segmented manually by two expert radiologists. Univariate analysis, linear regression modelling and pattern recognition methods were used as statistical and classification procedures.

RESULTS

The best results at univariate analysis were reached by the wavelet_LLH_glcm_JointEntropy extracted by T2W SPACE sequence with accuracy of 92%. Linear regression model increased the performance obtained respect to the univariate analysis. The best results were obtained by a linear regression model of 15 significant features extracted by the T2W SPACE sequence with accuracy of 94%, a sensitivity of 92% and a specificity of 95%. The best classifier among the tested pattern recognition approaches was k-nearest neighbours (KNN); however, KNN achieved lower precision than the best linear regression model.

CONCLUSIONS

Radiomics metrics allow the mucinous subtype lesion characterization, in order to obtain a more personalized approach. We demonstrated that the best performance was obtained by T2-W extracted textural metrics.

Radiomics and Machine Learning Analysis Based on Magnetic Resonance Imaging in the Assessment of Colorectal Liver Metastases Growth Pattern

To assess Radiomics and Machine Learning Analysis in Liver Colon and Rectal Cancer Metastases (CRLM) Growth Pattern, we evaluated, retrospectively, a training set of 51 patients with 121 liver metastases and an external validation set of 30 patients with a single lesion. All patients were subjected to MRI studies in pre-surgical setting. For each segmented volume of interest (VOI), 851 radiomics features were extracted using PyRadiomics package. Nonparametric test, univariate, linear regression analysis and patter recognition approaches were performed. The best results to discriminate expansive versus infiltrative front of tumor growth with the highest accuracy and AUC at univariate analysis were obtained by the wavelet_LHH_glrlm_ShortRunLowGray Level Emphasis from portal phase of contrast study. With regard to linear regression model, this increased the performance obtained respect to the univariate analysis for each sequence except that for EOB-phase sequence. The best results were obtained by a linear regression model of 15 significant features extracted by the T2-W SPACE sequence. Furthermore, using pattern recognition approaches, the diagnostic performance to discriminate the expansive versus infiltrative front of tumor growth increased again and the best classifier was a weighted KNN trained with the 9 significant metrics extracted from the portal phase of contrast study, with an accuracy of 92% on training set and of 91% on validation set. In the present study, we have demonstrated as Radiomics and Machine Learning Analysis, based on EOB-MRI study, allow to identify several biomarkers that permit to recognise the different Growth Patterns in CRLM.

Radiomics textural features by MR imaging to assess clinical outcomes following liver resection in colorectal liver metastases

PURPOSE

To assess the efficacy of radiomics features obtained by T2-weighted sequences to predict clinical outcomes following liver resection in colorectal liver metastases patients.

METHODS

This retrospective analysis was approved by the local Ethical Committee board and radiological databases were interrogated, from January 2018 to May 2021, to select patients with liver metastases with pathological proof and MRI study in pre-surgical setting. The cohort of patients included a training set and an external validation set. The internal training set included 51 patients with 61 years of median age and 121 liver metastases. The validation cohort consisted a total of 30 patients with single lesion with 60 years of median age. For each volume of interest, 851 radiomics features were extracted as median values using PyRadiomics. Nonparametric test, intraclass correlation, receiver operating characteristic (ROC) analysis, linear regression modelling and pattern recognition methods (support vector machine (SVM), k-nearest neighbours (KNN), artificial neural network (NNET) and decision tree (DT)) were considered.

RESULTS

The best predictor to discriminate expansive versus infiltrative front of tumour growth was obtained by wavelet_LHL_gldm_DependenceNonUniformityNormalized with an accuracy of 82%; to discriminate high grade versus low grade or absent was the wavelet_LLH_glcm_Imc1 with accuracy of 88%; to differentiate the mucinous type of tumour was the wavelet_LLH_glcm_JointEntropy with accuracy of 92% while to identify tumour recurrence was the wavelet_LLL_glcm_Correlation with accuracy of 85%. Linear regression model increased the performance obtained with respect to the univariate analysis exclusively in the discrimination of expansive versus infiltrative front of tumour growth reaching an accuracy of 90%, a sensitivity of 95% and a specificity of 80%. Considering significant texture metrics tested with pattern recognition approaches, the best performance was reached by the KNN in the discrimination of the tumour budding considering the four textural predictors obtaining an accuracy of 93%, a sensitivity of 81% and a specificity of 97%.

CONCLUSIONS

Ours results confirmed the capacity of radiomics to identify as biomarkers, several prognostic features that could affect the treatment choice in patients with liver metastases, in order to obtain a more personalized approach.

An update on radiomics techniques in primary liver cancers

BACKGROUND

Radiomics is a progressing field of research that deals with the extraction of quantitative metrics from medical images. Radiomic features detention indirectly tissue features such as heterogeneity and shape and can, alone or in combination with demographic, histological, genomic, or proteomic data, be used for decision support system in clinical setting.

METHODS

This article is a narrative review on Radiomics in Primary Liver Cancers. Particularly, limitations and future perspectives are discussed.

RESULTS

In oncology, assessment of tissue heterogeneity is of particular interest: genomic analysis have demonstrated that the degree of tumour heterogeneity is a prognostic determinant of survival and an obstacle to cancer control. Therefore, that Radiomics could support cancer detection, diagnosis, evaluation of prognosis and response to treatment, so as could supervise disease status in hepatocellular carcinoma (HCC) and Intrahepatic Cholangiocarcinoma (ICC) patients. Radiomic analysis is a convenient radiological image analysis technique used to support clinical decisions as it is able to provide prognostic and / or predictive biomarkers that allow a fast, objective and repeatable tool for disease monitoring.

CONCLUSIONS

Although several studies have shown that this analysis is very promising, there is little standardization and generalization of the results, which limits the translation of this method into the clinical context. The limitations are mainly related to the evaluation of data quality, repeatability, reproducibility, overfitting of the model.

TRIAL REGISTRATION

Not applicable.

EOB-MR Based Radiomics Analysis to Assess Clinical Outcomes following Liver Resection in Colorectal Liver Metastases

Simple Summary

The aim of this study was to assess the efficacy of radiomics features obtained by EOB-MRI phase in order to predict clinical outcomes following liver resection in Colorectal Liver Metastases Patients, and evaluate recurrence, mutational status, pathological characteristic (mucinous) and surgical resection margin. Ours results confirmed the capacity of radiomics to identify, as biomarkers, several prognostic features that could affect the treatment choice in patients with liver metastases, in order to obtain a more personalized approach. These results were confirmed by external validation dataset. We obtained a good performance considering the single textural significant metric in the identification of front of tumor growth (expansive versus infiltrative) and tumor budding (high grade versus low grade or absent), in the recognition of mucinous type and in the detection of recurrences.

Abstract

The aim of this study was to assess the efficacy of radiomics features obtained by EOB-MRI phase in order to predict clinical outcomes following liver resection in Colorectal Liver Metastases Patients, and evaluate recurrence, mutational status, pathological characteristic (mucinous) and surgical resection margin. This retrospective analysis was approved by the local Ethical Committee board of National Cancer of Naples, IRCCS “Fondazione Pascale”. Radiological databases were interrogated from January 2018 to May 2021 in order to select patients with liver metastases with pathological proof and EOB-MRI study in pre-surgical setting. The cohort of patients included a training set (51 patients with 61 years of median age and 121 liver metastases) and an external validation set (30 patients with single lesion with 60 years of median age). For each segmented volume of interest by 2 expert radiologists, 851 radiomics features were extracted as median values using PyRadiomics. non-parametric test, intraclass correlation, receiver operating characteristic (ROC) analysis, linear regression modelling and pattern recognition methods (support vector machine (SVM), k-nearest neighbors (KNN), artificial neural network (NNET), and decision tree (DT)) were considered. The best predictor to discriminate expansive versus infiltrative front of tumor growth was HLH_glcm_MaximumProbability extraxted on VIBE_FA30 with an accuracy of 84%, a sensitivity of 83%, and a specificity of 82%. The best predictor to discriminate tumor budding was Inverse Variance obtained by the original GLCM matrix extraxted on VIBE_FA30 with an accuracy of 89%, a sensitivity of 96% and a specificity of 65%. The best predictor to differentiate the mucinous type of tumor was the HHL_glszm_ZoneVariance extraxted on VIBE_FA30 with an accuracy of 85%, a sensitivity of 46% and a specificity of 95%. The best predictor to identify tumor recurrence was the LHL_glcm_Correlation extraxted on VIBE_FA30 with an accuracy of 86%, a sensitivity of 52% and a specificity of 97%. The best linear regression model was obtained in the identification of the tumor growth front considering the height textural significant metrics by VIBE_FA10 (an accuracy of 89%; sensitivity of 93% and a specificity of 82%). Considering significant texture metrics tested with pattern recognition approaches, the best performance for each outcome was reached by a KNN in the identification of recurrence with the 3 textural significant features extracted by VIBE_FA10 (AUC of 91%, an accuracy of 93%; sensitivity of 99% and a specificity of 77%). Ours results confirmed the capacity of radiomics to identify as biomarkers, several prognostic features that could affect the treatment choice in patients with liver metastases, in order to obtain a more personalized approach.

Contrast MR-Based Radiomics and Machine Learning Analysis to Assess Clinical Outcomes following Liver Resection in Colorectal Liver Metastases: A Preliminary Study

Simple Summary

The objective of the study was to evaluate the radiomics features obtained by contrast MRI studies as prognostic biomarkers in colorectal liver metastases patients to predict clinical outcomes following liver resection. We demonstrated a good performance considering the single textural significant metric in the identification of front of tumor growth (expansive versus infiltrative) and tumor budding (high grade versus low grade or absent), in the recognition of mucinous type and in the detection of recurrences. Moreover, considering linear regression models or neural network classifiers in a multivariate approach was possible to increase the performance in terms of accuracy, sensitivity, and specificity.

Abstract

Purpose: To assess radiomics features efficacy obtained by arterial and portal MRI phase in the prediction of clinical outcomes in the colorectal liver metastases patients, evaluating recurrence, mutational status, pathological characteristic (mucinous and tumor budding) and surgical resection margin. Methods: This retrospective analysis was approved by the local Ethical Committee board, and radiological databases were used to select patients with colorectal liver metastases with pathological proof and MRI study in a pre-surgical setting after neoadjuvant chemotherapy. The cohort of patients included a training set (51 patients with 61 years of median age and 121 liver metastases) and an external validation set (30 patients with single lesion with 60 years of median age). For each segmented volume of interest on MRI by two expert radiologists, 851 radiomics features were extracted as median values using the PyRadiomics tool. Non-parametric Kruskal-Wallis test, intraclass correlation, receiver operating characteristic (ROC) analysis, linear regression modelling and pattern recognition methods (support vector machine (SVM), k-nearest neighbors (KNN), artificial neural network (NNET), and decision tree (DT)) were considered. Results: The best predictor to discriminate expansive versus infiltrative tumor growth front was wavelet_LHH_glrlm_ShortRunLowGrayLevelEmphasis extracted on portal phase with accuracy of 82%, sensitivity of 84%, and specificity of 77%. The best predictor to discriminate tumor budding was wavelet_LLH_firstorder_10Percentile extracted on portal phase with accuracy of 92%, a sensitivity of 96%, and a specificity of 81%. The best predictor to differentiate the mucinous type of tumor was the wavelet_LLL_glcm_ClusterTendency extracted on portal phase with accuracy of 88%, a sensitivity of 38%, and a specificity of 100%. The best predictor to identify the recurrence was the wavelet_HLH_ngtdm_Complexity extracted on arterial phase with accuracy of 90%, a sensitivity of 71%, and a specificity of 95%. The best linear regression model was obtained in the identification of mucinous type considering the 13 textural significant metrics extracted by arterial phase (accuracy of 94%, sensitivity of 77% and a specificity of 99%). The best results were obtained in the identification of tumor budding with the eleven textural significant features extracted by arterial phase using a KNN (accuracy of 95%, sensitivity of 84%, and a specificity of 99%). Conclusions: Our results confirmed the capacity of radiomics to identify as biomarkers and several prognostic features that could affect the treatment choice in patients with liver metastases in order to obtain a more personalized approach.