Efficacy of MRI data harmonization in the age of machine learning: a multicenter study across 36 datasets.

Pooling publicly-available MRI data from multiple sites allows to assemble extensive groups of subjects, increase statistical power, and promote data reuse with machine learning techniques. The harmonization of multicenter data is necessary to reduce the confounding effect associated with non-biological sources of variability in the data. However, when applied to the entire dataset before machine learning, the harmonization leads to data leakage, because information outside the training set may affect model building, and potentially falsely overestimate performance. We propose a 1) measurement of the efficacy of data harmonization; 2) harmonizer transformer, i.e., an implementation of the ComBat harmonization allowing its encapsulation among the preprocessing steps of a machine learning pipeline, avoiding data leakage by design. We tested these tools using brain T1-weighted MRI data from 1740 healthy subjects acquired at 36 sites. After harmonization, the site effect was removed or reduced, and we showed the data leakage effect in predicting individual age from MRI data, highlighting that introducing the harmonizer transformer into a machine learning pipeline allows for avoiding data leakage by design.

Impact of seed amplification assay and surface-enhanced Raman spectroscopy combined approach on the clinical diagnosis of Alzheimer's disease.

BACKGROUND: The current diagnosis of Alzheimer's disease (AD) is based on a series of analyses which involve clinical, instrumental and laboratory findings. However, signs, symptoms and biomarker alterations observed in AD might overlap with other dementias, resulting in misdiagnosis. METHODS: Here we describe a new diagnostic approach for AD which takes advantage of the boosted sensitivity in biomolecular detection, as allowed by seed amplification assay (SAA), combined with the unique specificity in biomolecular recognition, as provided by surface-enhanced Raman spectroscopy (SERS). RESULTS: The SAA-SERS approach supported by machine learning data analysis allowed efficient identification of pathological Aß oligomers in the cerebrospinal fluid of patients with a clinical diagnosis of AD or mild cognitive impairment due to AD. CONCLUSIONS: Such analytical approach can be used to recognize disease features, thus allowing early stratification and selection of patients, which is fundamental in clinical treatments and pharmacological trials.

Collinearity and Dimensionality Reduction in Radiomics: Effect of Preprocessing Parameters in Hypertrophic Cardiomyopathy Magnetic Resonance T1 and T2 Mapping.

Radiomics and artificial intelligence have the potential to become a valuable tool in clinical applications. Frequently, radiomic analyses through machine learning methods present issues caused by high dimensionality and multicollinearity, and redundant radiomic features are usually removed based on correlation analysis. We assessed the effect of preprocessing-in terms of voxel size resampling, discretization, and filtering-on correlation-based dimensionality reduction in radiomic features from cardiac T1 and T2 maps of patients with hypertrophic cardiomyopathy. For different combinations of preprocessing parameters, we performed a dimensionality reduction of radiomic features based on either Pearson's or Spearman's correlation coefficient, followed by the computation of the stability index. With varying resampling voxel size and discretization bin width, for both T1 and T2 maps, Pearson's and Spearman's dimensionality reduction produced a slightly different percentage of remaining radiomic features, with a relatively high stability index. For different filters, the remaining features' stability was instead relatively low. Overall, the percentage of eliminated radiomic features through correlation-based dimensionality reduction was more dependent on resampling voxel size and discretization bin width for textural features than for shape or first-order features. Notably, correlation-based dimensionality reduction was less sensitive to preprocessing when considering radiomic features from T2 compared with T1 maps.

NAVIGATOR: an Italian regional imaging biobank to promote precision medicine for oncologic patients.

NAVIGATOR is an Italian regional project boosting precision medicine in oncology with the aim of making it more predictive, preventive, and personalised by advancing translational research based on quantitative imaging and integrative omics analyses. The project's goal is to develop an open imaging biobank for the collection and preservation of a large amount of standardised imaging multimodal datasets, including computed tomography, magnetic resonance imaging, and positron emission tomography data, together with the corresponding patient-related and omics-related relevant information extracted from regional healthcare services using an adapted privacy-preserving model. The project is based on an open-source imaging biobank and an open-science oriented virtual research environment (VRE). Available integrative omics and multi-imaging data of three use cases (prostate cancer, rectal cancer, and gastric cancer) will be collected. All data confined in NAVIGATOR (i.e., standard and novel imaging biomarkers, non-imaging data, health agency data) will be used to create a digital patient model, to support the reliable prediction of the disease phenotype and risk stratification. The VRE that relies on a well-established infrastructure, called D4Science.org, will further provide a multiset infrastructure for processing the integrative omics data, extracting specific radiomic signatures, and for identification and testing of novel imaging biomarkers through big data analytics and artificial intelligence.

Image resampling and discretization effect on the estimate of myocardial radiomic features from T1 and T2 mapping in hypertrophic cardiomyopathy.

Radiomics is emerging as a promising and useful tool in cardiac magnetic resonance (CMR) imaging applications. Accordingly, the purpose of this study was to investigate, for the first time, the effect of image resampling/discretization and filtering on radiomic features estimation from quantitative CMR T1 and T2 mapping. Specifically, T1 and T2 maps of 26 patients with hypertrophic cardiomyopathy (HCM) were used to estimate 98 radiomic features for 7 different resampling voxel sizes (at fixed bin width), 9 different bin widths (at fixed resampling voxel size), and 7 different spatial filters (at fixed resampling voxel size/bin width). While we found a remarkable dependence of myocardial radiomic features from T1 and T2 mapping on image filters, many radiomic features showed a limited sensitivity to resampling voxel size/bin width, in terms of intraclass correlation coefficient (> 0.75) and coefficient of variation (< 30%). The estimate of most textural radiomic features showed a linear significant (p < 0.05) correlation with resampling voxel size/bin width. Overall, radiomic features from T2 maps have proven to be less sensitive to image preprocessing than those from T1 maps, especially when varying bin width. Our results might corroborate the potential of radiomics from T1/T2 mapping in HCM and hopefully in other myocardial diseases.

Paper-based genetic assays with bioconjugated gold nanorods and an automated readout pipeline.

Paper-based biosensors featuring immunoconjugated gold nanoparticles have gained extraordinary momentum in recent times as the platform of choice in key cases of field applications, including the so-called rapid antigen tests for SARS-CoV-2. Here, we propose a revision of this format, one that may leverage on the most recent advances in materials science and data processing. In particular, we target an amplifiable DNA rather than a protein analyte, and we replace gold nanospheres with anisotropic nanorods, which are intrinsically brighter by a factor of ~ 10, and multiplexable. By comparison with a gold-standard method for dot-blot readout with digoxigenin, we show that gold nanorods entail much faster and easier processing, at the cost of a higher limit of detection (from below 1 to 10 ppm in the case of plasmid DNA containing a target transgene, in our current setup). In addition, we test a complete workflow to acquire and process photographs of dot-blot membranes with custom-made hardware and regression tools, as a strategy to gain more analytical sensitivity and potential for quantification. A leave-one-out approach for training and validation with as few as 36 sample instances already improves the limit of detection reached by the naked eye by a factor around 2. Taken together, we conjecture that the synergistic combination of new materials and innovative tools for data processing may bring the analytical sensitivity of paper-based biosensors to approach the level of lab-grade molecular tests.

A deep look into radiomics.

Radiomics is a process that allows the extraction and analysis of quantitative data from medical images. It is an evolving field of research with many potential applications in medical imaging. The purpose of this review is to offer a deep look into radiomics, from the basis, deeply discussed from a technical point of view, through the main applications, to the challenges that have to be addressed to translate this process in clinical practice. A detailed description of the main techniques used in the various steps of radiomics workflow, which includes image acquisition, reconstruction, pre-processing, segmentation, features extraction and analysis, is here proposed, as well as an overview of the main promising results achieved in various applications, focusing on the limitations and possible solutions for clinical implementation. Only an in-depth and comprehensive description of current methods and applications can suggest the potential power of radiomics in fostering precision medicine and thus the care of patients, especially in cancer detection, diagnosis, prognosis and treatment evaluation.

Artificial intelligence applications in medical imaging: A review of the medical physics research in Italy.

PURPOSE: To perform a systematic review on the research on the application of artificial intelligence (AI) to imaging published in Italy and identify its fields of application, methods and results. MATERIALS AND METHODS: A Pubmed search was conducted using terms Artificial Intelligence, Machine Learning, Deep learning, imaging, and Italy as affiliation, excluding reviews and papers outside time interval 2015-2020. In a second phase, participants of the working group AI4MP on Artificial Intelligence of the Italian Association of Physics in Medicine (AIFM) searched for papers on AI in imaging. RESULTS: The Pubmed search produced 794 results. 168 studies were selected, of which 122 were from Pubmed search and 46 from the working group. The most used imaging modality was MRI (44%) followed by CT(12%) ad radiography/mammography (11%). The most common clinical indication were neurological diseases (29%) and diagnosis of cancer (25%). Classification was the most common task for AI (57%) followed by segmentation (16%). 65% of studies used machine learning and 35% used deep learning. We observed a rapid increase of research in Italy on artificial intelligence in the last 5 years, peaking at 155% from 2018 to 2019. CONCLUSIONS: We are witnessing an unprecedented interest in AI applied to imaging in Italy, in a diversity of fields and imaging techniques. Further initiatives are needed to build common frameworks and databases, collaborations among different types of institutions, and guidelines for research on AI.

Machine and Deep Learning Prediction Of Prostate Cancer Aggressiveness Using Multiparametric MRI.

Prostate cancer (PCa) is the most frequent male malignancy and the assessment of PCa aggressiveness, for which a biopsy is required, is fundamental for patient management. Currently, multiparametric (mp) MRI is strongly recommended before biopsy. Quantitative assessment of mpMRI might provide the radiologist with an objective and noninvasive tool for supporting the decision-making in clinical practice and decreasing intra- and inter-reader variability. In this view, high dimensional radiomics features and Machine Learning (ML) techniques, along with Deep Learning (DL) methods working on raw images directly, could assist the radiologist in the clinical workflow. The aim of this study was to develop and validate ML/DL frameworks on mpMRI data to characterize PCas according to their aggressiveness. We optimized several ML/DL frameworks on T2w, ADC and T2w+ADC data, using a patient-based nested validation scheme. The dataset was composed of 112 patients (132 peripheral lesions with Prostate Imaging Reporting and Data System (PI-RADS) score ≥ 3) acquired following both PI-RADS 2.0 and 2.1 guidelines. Firstly, ML/DL frameworks trained and validated on PI-RADS 2.0 data were tested on both PI-RADS 2.0 and 2.1 data. Then, we trained, validated and tested ML/DL frameworks on a multi PI-RADS dataset. We reported the performances in terms of Area Under the Receiver Operating curve (AUROC), specificity and sensitivity. The ML/DL frameworks trained on T2w data achieved the overall best performance. Notably, ML and DL frameworks trained and validated on PI-RADS 2.0 data obtained median AUROC values equal to 0.750 and 0.875, respectively, on unseen PI-RADS 2.0 test set. Similarly, ML/DL frameworks trained and validated on multi PI-RADS T2w data showed median AUROC values equal to 0.795 and 0.750, respectively, on unseen multi PI-RADS test set. Conversely, all the ML/DL frameworks trained and validated on PI-RADS 2.0 data, achieved AUROC values no better than the chance level when tested on PI-RADS 2.1 data. Both ML/DL techniques applied on mpMRI seem to be a valid aid in predicting PCa aggressiveness. In particular, ML/DL frameworks fed with T2w images data (objective, fast and non-invasive) show good performances and might support decision-making in patient diagnostic and therapeutic management, reducing intra- and inter-reader variability.

Label-free SERS detection of proteins based on machine learning classification of chemo-structural determinants.

Establishing standardized methods for a consistent analysis of spectral data remains a largely underexplored aspect in surface-enhanced Raman spectroscopy (SERS), particularly applied to biological and biomedical research. Here we propose an effective machine learning classification of protein species with closely resembled spectral profiles by a mixed data processing based on principal component analysis (PCA) applied to multipeak fitting on SERS spectra. This strategy simultaneously assures a successful discrimination of proteins and a thorough characterization of the chemostructural differences among them, ultimately opening up new routes for SERS evolution toward sensing applications and diagnostics of interest in life sciences.

Ficoll as testing material for diffusion weighted imaging-quality assurance phantoms.

PURPOSE: The aim of this work is to test the use of aqueous solutions of Ficoll®**, a highly branched polymer displaying crowding properties, to build a phantom suitable for Diffusion Weighted Imaging (DWI) in Magnetic Resonance Imaging (MRI). METHODS: We developed a test object made of a cylindrical plastic container with a precise geometrical arrangement suitable for measuring several samples at the same time. The container was designed to host single vials with variable geometry and number, and to fit inside common commercial head coils for MRI scanners. In our experiments, vials were filled with 8 aqueous solutions of Ficoll 70 and Ficoll 400 spanning a range of polymer concentration from 5 to 30% by weight. Vials containing ultra-pure water were also used as reference. Experiments were performed on both 1.5 and 3 T clinical scanners (GE, Philips and Siemens), under the conditions of a standard clinical examination. RESULTS: The geometry of the phantom provided reduced imaging artifacts, especially image distortions at magnetic interfaces. We found that the Apparent Diffusion Coefficient (ADC) varied in the range of 0.00125-0.00223 mm2/s and decreased with Ficoll concentration. ADC vs Ficoll concentration exhibited a linear trend. Results were consistent over time and among different MRI clinical scanners, showing an average variability of 3% at 1.5 T and of 7.5% at 3 T. Moreover, no substantial difference was found between Ficoll 70 and 400. By varying Ficoll concentration, ADC can be modulated to approach tissue-mimicking values. Preliminary results for relaxation measurements proved that both T1 and T2 decreased with Ficoll concentration in the ranges 1.3-2.4 s and 150-800 ms respectively. CONCLUSIONS: In this work, we propose a 3D phantom design based on the widespread crowding agent Ficoll, which is suitable for DWI quality assurance purposes in MRI acquisitions. Aqueous Ficoll solutions provide good performance in terms of stability, ease of preparation, and safety.

Optically activated and interrogated plasmonic hydrogels for applications in wound healing.

We disclose the use of hybrid materials featuring Au/Ag core/shell nanorods in porous chitosan/polyvinyl alcohol scaffolds for applications in tissue engineering and wound healing. The combination of Au and Ag in a single construct provides synergistic opportunities for optical activation of functions as near infrared laser tissue bonding, and remote interrogation to return parameters of prognostic relevance in wound healing monitoring. In particular, the bimetallic component ensures optical tunability, enhanced shelf life and photothermal stability, serves as a reservoir of germicidal silver cations, and changes in near-infrared and visible color according to the environmental level of oxidative stress. At the same time, the polymeric blend is ideal to bind connective tissue upon photothermal activation, and to support fabrication processes that provide high porosity, such as electrospinning, thus putting all the premises for cellular repopulation and antimicrobial protection.

Comprehensive Analysis of Radiomic Datasets by RadAR.

Quantitative analysis of biomedical images, referred to as radiomics, is emerging as a promising approach to facilitate clinical decisions and improve patient stratification. The typical radiomic workflow includes image acquisition, segmentation, feature extraction, and analysis of high-dimensional datasets. While procedures for primary radiomic analyses have been established in recent years, processing the resulting radiomic datasets remains a challenge due to the lack of specific tools for doing so. Here we present RadAR (Radiomics Analysis with R), a new software to perform comprehensive analysis of radiomic features. RadAR allows users to process radiomic datasets in their entirety, from data import to feature processing and visualization, and implements multiple statistical methods for analysis of these data. We used RadAR to analyze the radiomic profiles of more than 850 patients with cancer from publicly available datasets and showed that it was able to recapitulate expected results. These results demonstrate RadAR as a reliable and valuable tool for the radiomics community. SIGNIFICANCE: A new computational tool performs comprehensive analysis of high-dimensional radiomic datasets, recapitulating expected results in the analysis of radiomic profiles of >850 patients with cancer from independent datasets.

Parametrical Optomechanical Oscillations in PhoXonic Whispering Gallery Mode Resonators.

We report on the experimental and theoretical analysis of parametrical optomechanical oscillations in hollow spherical phoxonic whispering gallery mode resonators due to radiation pressure. The optically excited acoustic eigenmodes of the phoxonic cavity oscillate regeneratively leading to parametric oscillation instabilities.

Selective coupling of Whispering Gallery Modes in film coated micro-resonators.

Whispering Gallery Mode (WGM) micro-resonators like microspheres or microtoroids are typically used as high-Q cavity substrate on which a functional film coating is deposited. In order to exploit the coating properties a critical step is the efficient excitation of WGMs mainly contained inside the deposited layer. We developed a simple method able to assess whether or not these modes are selectively excited. The method is based on monitoring the thermal shift of the excited resonance, which uniquely depends on the thermo-optic coefficient and on the thermal expansion coefficient of the material in which the mode is embedded.

Optical Microbubble Resonators with High Refractive Index Inner Coating for Bio-Sensing Applications: An Analytical Approach.

The design of Whispering Gallery Mode Resonators (WGMRs) used as an optical transducer for biosensing represents the first and crucial step towards the optimization of the final device performance in terms of sensitivity and Limit of Detection (LoD). Here, we propose an analytical method for the design of an optical microbubble resonator (OMBR)-based biosensor. In order to enhance the OMBR sensing performance, we consider a polymeric layer of high refractive index as an inner coating for the OMBR. The effect of this layer and other optical/geometrical parameters on the mode field distribution, sensitivity and LoD of the OMBR is assessed and discussed, both for transverse electric (TE) and transverse magnetic (TM) polarization. The obtained results do provide physical insights for the development of OMBR-based biosensor.

Resonance Frequency of Optical Microbubble Resonators: Direct Measurements and Mitigation of Fluctuations.

This work shows the improvements in the sensing capabilities and precision of an Optical Microbubble Resonator due to the introduction of an encaging poly(methyl methacrylate) (PMMA) box. A frequency fluctuation parameter σ was defined as a score of resonance stability and was evaluated in the presence and absence of the encaging system and in the case of air- or water-filling of the cavity. Furthermore, the noise interference introduced by the peristaltic and the syringe pumping system was studied. The measurements showed a reduction of σ in the presence of the encaging PMMA box and when the syringe pump was used as flowing system.

Generation of hyper-parametric oscillations in silica microbubbles.

Cavity resonant enhanced stimulated Raman scattering (SRS), four-wave mixing, and broadband hyper-parametric oscillation in silica microbubble whispering gallery mode resonators (WGMR) in forward and backward directions are reported in this Letter. We show that microbubbles can operate not only in a highly ideal two-photon emission regime, but also generate combs, both natively and multi-mode spaced. The nonlinear process is phase matched because of the interaction of different mode families of the resonator.

Confocal reflectance microscopy for determination of microbubble resonator thickness.

Optical Micro Bubble Resonators (OMBR) are emerging as new type of sensors characterized by high Q-factor and embedded micro-fluidic. Sensitivity is related to cavity field penetration and, therefore, to the resonator thickness. At the state of the art, methods for OMBR's wall thickness evaluation rely only on a theoretical approach. The purpose of this study is to create a non-destructive method for measuring the shell thickness of a microbubble using reflectance confocal microscopy. The method was validated through measurements on etched capillaries with different thickness and finally it was applied on microbubble resonators.