Extraction Of Quantitative Features Research Articles

Radiomic Applications in Skull Base Pathology: A Systematic Review of Potential Clinical Uses

Objectives Radiomics involves the extraction and analysis of numerous quantitative features of medical imaging which can add more information from radiological images often beyond initial comprehension of a clinician. Unlike deep learning, radiomics allows some understanding of identified quantitative features for clinical prediction. We sought to explore the current state of radiomics applications in the skull base literature. Methods A systematic review of studies evaluating radiomics in skull base was performed, including those with and without machine-learning approaches. Studies were summarized into thematic elements as well as specific pathologies. Results A total of 102 studies with 26,280 radiographic images were included. The earliest radiomic study was published in 2017 with exponential growth in research since then. Most studies focused on tumor diagnosis (40.8%), followed by tumor prognosis (31.1%), automated segmentation (16.5%), other applications (7.8%), and lastly prediction of intraoperative features (3.9%). Pituitary adenomas (41.7%) and vestibular schwannomas (18.4%) represented the most commonly evaluated pathologies; however, radiomics could be applied to a heterogeneous collection of skull base pathologies. The average study included 258677 cases (range 4; 6755). Conclusion Radiomics offers many functions in treating skull base pathology and will likely be an essential component of future clinical care. Larger sample sizes, validation of predictive models, and clinical application are needed. Further investigation into the strengths and weaknesses of radiomic applications in skull base treatments is warranted.

Security monitoring via sound analysis and voice identification with artificial intelligence

The article demonstrates the possibility of monitoring user access through authentication based on voice profiles using the means of Artificial Intelligence. A two-stage approach is proposed for sound analysis and voice recognition using Feed-Forward Neural Networks (FFNNs) and Cascade-Forward Neural Networks (CFNNs). Seven test voice profiles were pre-processed to extract quantitative sound features. The procedure involves registration of a set of sound parameters concerning three categories, respectively, for all audio and acoustic measurements in the entire sound spectrum, measurements up to and above 100 dBA. The neural architectures were trained with Scaled Gradient Descent (SCG) and Levenberg Marquardt (LM) algorithms, using different transfer functions in the output structural layers. In the initial phase of neural training, the entire sound spectrum of registered indicators was used, and high levels of Accuracy around 90.0% were reached. Subsequently, steps were taken to reduce the informative features when searching for similar levels of accuracy in order to limit the necessary computational procedures in neural training, but maintain the threshold of successful user authentication. In the analysis of neural performance, in addition to accuracy, additional criteria were used, namely Mean-Squared Error (MSE) and Root Mean Squared Error (RMSE). About the achieved and analyzed results, a synthesis was conducted of a set of four informative features with the highest significance, respectively LAE (A-weighted, sound exposure level), Laeq (A-weighted, equivalent sound level), LAF (A-weighted, fast time-constant, sound level) and LAS (A-weighted, slow time constant response, sound level). In the course of subsequent neural training processes, unsuitability was found when using the Log-sigmoid activation type with greatly underestimated accuracy readings and errors below 58.0% and above thresholds of 0.2300 and 0.4800. Positive performance indicators of voice recognition were achieved with Softmax and Hyperbolic tangent sigmoid activations in SCG and LM training procedures in levels of accuracy of 98.7 % and 96.1 % at FFNN models. Successful correct recognition of the test voice profiles on access and security personalization with a quantitative equivalent of 100.0 % accuracy was achieved in the Linear transfer function for Cascade-Forward Neural Networks. The proposed method and the synthesized neural models in the research can be used as units and modules in access control systems with biometric diagnostics and intelligent recognition of employees in company departments to electronically store classified information and physical access control.

Abstract PO3-07-04: Prediction of PAM50 molecular subtypes from H&E-stained breast cancer specimens using tumor microenvironment features and additive multiple instance learning models

Abstract Background: PAM50, a 50-gene signature, classifies breast cancers into one of five subtypes (basal, luminal A, luminal B, HER2-enriched, and normal-like), revealing information about underlying tumor biology, and has emerged as a key prognostic indicator influencing treatment decisions. There is growing interest in bridging the gap between expression-based metrics and histopathology, where immunohistochemistry (IHC) and sequencing-based approaches have been proposed for this purpose. However, hematoxylin and eosin (H&E)-stained slides are ubiquitously utilized by pathologists for cancer diagnosis, while IHC and sequencing-based approaches require additional tissue and specialized processing and/or analysis. Here, we describe a computer vision-based approach to predict PAM50 classification using H&E-stained whole slide images (WSIs). Methods: We obtained expression-based PAM50 subtype labels and corresponding H&E-stained WSIs for 961 breast carcinomas from the TCGA BRCA cohort. We used two separate machine learning (ML) approaches to predict PAM50 subtypes from WSIs. In the first approach, we deployed previously trained PathExplore models to extract quantitative human-interpretable features (HIFs) that summarize the TME. We subsequently trained random forest classification models on these HIFs to predict PAM50 subtypes. For the second approach, we developed additive multiple instance learning (aMIL) models. Additionally, we explored the effects of PAM50 subtype labeling and aggregation strategies beyond the 5-class approach. Our 3-class approach combines Luminal A and B, as seen in IHC efforts to increase agreement with PAM50 assays, while excluding Normal, a category containing few and heterogeneous samples. We also performed binary classification for each subtype in the 3-class model (e.g. luminal vs. other). Slides were split into training (60%), validation (20%), and test (20%) sets, stratified by PAM50 labels, and model performance was assessed using the area under the receiver operator curve (AUROC) metric on the held-out test set, using a one vs. rest approach for multi-class models. To establish a baseline for PAM50 prediction, we developed random forest classification models using only clinical covariates (tumor stage, histologic grade, histological subtype, and BRCA1/2 status). Results: We compared the performance of our two ML models (HIF and aMIL) to that of the baseline model, and we report the AUROC values in Table 1. These models both performed well in predicting Basal, Luminal A, Luminal B, and Luminal (A+B), while the model performance was less strong for predictions of the HER2 and Normal classifications. The three-class model showed improved performance of predicting Luminal classifications relative to the five-class model that separates Luminal A and B. Although simplifying classification problems to a binary use case typically provides improved performance, this phenomenon was not observed for any of the PAM50 subtypes. Conclusions: These results demonstrate that AI-powered digital pathology can accurately and reproducibly perform molecular-based classification tasks, such as predicting PAM50 classifications, using WSIs, suggesting a more efficient path toward clinically relevant breast cancer characterization. Table 1. Performance of all models in predicting PAM50 molecular subtypes. AUROC values are shown. Shaded cells represent the best test-set performance for each class (row). Citation Format: Maria Guramare, Syed Ashar Javed, Christian Kirkup, Dinkar Juyal, Jacqueline Brosnan-Cashman, Victoria Mountain, Ryan Leung, Bahar Rahsepar, John Abel, Amaro Taylor-Weiner, Jake Conway. Prediction of PAM50 molecular subtypes from H&E-stained breast cancer specimens using tumor microenvironment features and additive multiple instance learning models [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO3-07-04.

MF-Net: multi-scale feature extraction-integration network for unsupervised deformable registration.

Deformable registration plays a fundamental and crucial role in scenarios such as surgical navigation and image-assisted analysis. While deformable registration methods based on unsupervised learning have shown remarkable success in predicting displacement fields with high accuracy, many existing registration networks are limited by the lack of multi-scale analysis, restricting comprehensive utilization of global and local features in the images. To address this limitation, we propose a novel registration network called multi-scale feature extraction-integration network (MF-Net). First, we propose a multiscale analysis strategy that enables the model to capture global and local semantic information in the image, thus facilitating accurate texture and detail registration. Additionally, we introduce grouped gated inception block (GI-Block) as the basic unit of the feature extractor, enabling the feature extractor to selectively extract quantitative features from images at various resolutions. Comparative experiments demonstrate the superior accuracy of our approach over existing methods.

Single and multiple scattering quantitative ultrasound methods in muscle

Increased intramuscular fat—or fatty infiltration (FI)—often develops as a consequence of various musculoskeletal and neuromuscular disorders. FI is an important factor for the assessment and intervention of surgical candidacy. Therefore, there is a need for an accessible, non-invasive quantitative, and objective evaluation of FI in muscle. We can exploit the complexity of ultrasound wave propagation in heterogeneous media to extract quantitative features of the fat distribution in muscle. From finite-difference time-domain simulations conducted with maps created from magnetic resonance imaging scans of healthy and injured shoulders, we acquire radiofrequency (RF) ultrasound data. Single and multiple scattering (SS and MS) components of the RF data are separated by using singular value decomposition and eigenvalue thresholding. Spectral and envelope quantitative ultrasound (QUS) parameters are computed from the SS component. MS-based QUS parameters are obtained by extracting the diffusion constant and tracking the SS intensity decay rate. We investigate the relationship between SS and MS-based QUS parameters and an increasing FI, because more FI will lead to more multiple scattering and modified ultrasonic signatures.

Machine learning-assisted high-content imaging analysis of 3D MCF7 microtissues for estrogenic effect prediction

Endocrine-disrupting chemicals (EDCs) pose a significant threat to human well-being and the ecosystem. However, in managing the many thousands of uncharacterized chemical entities, the high-throughput screening of EDCs using relevant biological endpoints remains challenging. Three-dimensional (3D) culture technology enables the development of more physiologically relevant systems in more realistic biochemical microenvironments. The high-content and quantitative imaging techniques enable quantifying endpoints associated with cell morphology, cell–cell interaction, and microtissue organization. In the present study, 3D microtissues formed by MCF-7 breast cancer cells were exposed to the model EDCs estradiol (E2) and propyl pyrazole triol (PPT). A 3D imaging and image analysis pipeline was established to extract quantitative image features from estrogen-exposed microtissues. Moreover, a machine-learning classification model was built using estrogenic-associated differential imaging features. Based on 140 common differential image features found between the E2 and PPT group, the classification model predicted E2 and PPT exposure with AUC-ROC at 0.9528 and 0.9513, respectively. Deep learning-assisted analysis software was developed to characterize microtissue gland lumen formation. The fully automated tool can accurately characterize the number of identified lumens and the total luminal volume of each microtissue. Overall, the current study established an integrated approach by combining non-supervised image feature profiling and supervised luminal volume characterization, which reflected the complexity of functional ER signaling and highlighted a promising conceptual framework for estrogenic EDC risk assessment.

Quantitation of Oncologic Image Features for Radiomic Analyses in PET.

Radiomics is an emerging and exciting field of study involving the extraction of many quantitative features from radiographic images. Positron emission tomography (PET) images are used in cancer diagnosis and staging. Utilizing radiomics on PET images can better quantify the spatial relationships between image voxels and generate more consistent and accurate results for diagnosis, prognosis, treatment, etc. This chapter gives the general steps a researcher would take to extract PET radiomic features from medical images and properly develop models to implement.

A Novel Feature Engineering Method Based on Latent Representation Learning for Radiomics: Application in NSCLC Subtype Classification.

Radiomics refers to the high-throughput extraction of quantitative features from medical images, and is widely used to construct machine learning models for the prediction of clinical outcomes, while feature engineering is the most important work in radiomics. However, current feature engineering methods fail to fully and effectively utilize the heterogeneity of features when dealing with different kinds of radiomics features. In this work, latent representation learning is first presented as a novel feature engineering approach to reconstruct a set of latent space features from original shape, intensity and texture features. This proposed method projects features into a subspace called latent space, in which the latent space features are obtained by minimizing a unique hybrid loss function including a clustering-like loss and a reconstruction loss. The former one ensures the separability among each class while the latter one narrows the gap between the original features and latent space features. Experiments were performed on a multi-center non-small cell lung cancer (NSCLC) subtype classification dataset from 8 international open databases. Results showed that compared with four traditional feature engineering methods (baseline, PCA, Lasso and L2,1-norm minimization), latent representation learning could significantly improve the classification performance of various machine learning classifiers on the independent test set (all p<0.001). Further on two additional test sets, latent representation learning also showed a significant improvement in generalization performance. Our research shows that latent representation learning is a more effective feature engineering method, which has the potential to be used as a general technology in a wide range of radiomics researches.

Radiomic features of the hippocampal based on magnetic resonance imaging in the menopausal mouse model linked to neuronal damage and cognitive deficits

Estrogen deficiency in the early postmenopausal phase is associated with an increased long-term risk of cognitive decline or dementia. Non-invasive characterization of the pathological features of the pathological hallmarks in the brain associated with postmenopausal women (PMW) could enhance patient management and the development of therapeutic strategies. Radiomics is a means to quantify the radiographic phenotype of a diseased tissue via the high-throughput extraction and mining of quantitative features from images acquired from modalities such as CT and magnetic resonance imaging (MRI). This study set out to explore the correlation between radiomics features based on MRI and pathological features of the hippocampus and cognitive function in the PMW mouse model. Ovariectomized (OVX) mice were used as PWM models. MRI scans were performed two months after surgery. The brain’s hippocampal region was manually annotated, and the radiomic features were extracted with PyRadiomics. Chemiluminescence was used to evaluate the peripheral blood estrogen level of mice, and the Morris water maze test was used to evaluate the cognitive ability of mice. Nissl staining and immunofluorescence were used to quantify neuronal damage and COX1 expression in brain sections of mice. The OVX mice exhibited marked cognitive decline, brain neuronal damage, and increased expression of mitochondrial complex IV subunit COX1, which are pathological phenomena commonly observed in the brains of AD patients, and these phenotypes were significantly correlated with radiomics features (p < 0.05, |r|>0.5), including Original_firstorder_Interquartile Range, Original_glcm_Difference Average, Original_glcm_Difference Average and Wavelet-LHH_glszm_Small Area Emphasis. Meanwhile, the above radiomics features were significantly different between the sham-operated and OVX groups (p < 0.01) and were associated with decreased serum estrogen levels (p < 0.05, |r|>0.5). This initial study indicates that the above radiomics features may have a role in the assessment of the pathology of brain damage caused by estrogen deficiency using routinely acquired structural MR images.

Radiomics to Differentiate Malignant and Benign Breast Lesions: A Systematic Review and Diagnostic Test Accuracy Meta-Analysis.

Breast cancer is a prevalent global health concern, necessitating accurate diagnostic tools for effective management. Diagnostic imaging plays a pivotal role in breast cancer diagnosis, staging, treatment planning, and outcome evaluation. Radiomics is an emerging field of study in medical imaging that contains a broad set of computational methods to extract quantitative features from radiographic images. This can be utilized to guide diagnosis, treatment response, and prognosis in clinical settings. A systematic review was performed in concordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and the Cochrane Handbook for Systematic Reviews of Diagnostic Test Accuracy. Quality was assessed using the radiomics quality score. Diagnostic sensitivity and specificity of radiomics analysis, with 95% confidence intervals (CIs), were included for meta-analysis. The area under the curve analysis was recorded. An extensive statistical analysis was performed following the Cochrane guidelines. Statistical significance was determined if p-values were less than 0.05. Statistical analyses were conducted using Review Manager (RevMan), Version 5.4.1. A total of 31 manuscripts involving 8,773 patients were included, with 17 contributing to the meta-analysis. The cohort comprised 56.2% malignant breast cancers and 43.8% benign breast lesions. MRI demonstrated a sensitivity of 0.91 (95% CI: 0.89-0.92) and a specificity of 0.84 (95% CI: 0.82-0.86) in differentiating between benign and malignant breast cancers. Mammography-based radiomic features predicted breast cancer subtype with a sensitivity of 0.79 (95% CI: 0.76-0.82) and a specificity of 0.81 (95% CI: 0.79-0.84). Ultrasound-based analysis yielded a sensitivity of 0.92 (95% CI: 0.90-0.94) and a specificity of 0.85 (95% CI: 0.83-0.88). Only one study reported the results of radiomic evaluation from CT, which had a sensitivity of 0.95 (95% CI: 0.88-0.99) and a specificity of 0.56 (95% CI: 0.45-0.67). Across different imaging modalities, radiomics exhibited robust diagnostic accuracy in differentiating benign and malignant breast lesions. The results underscore the potential of radiomic assessment as a minimally invasive alternative or adjunctive diagnostic tool for breast cancer. This is pioneering data that reports on a novel diagnostic approach that is understudied and underreported. However, due to study limitations, the complexity of this technology, and the need for future development, biopsy still remains the current gold standard method of determining breast cancer type.

Radiomics could predict surgery at 10 years in Crohn's disease.

Predicting clinical outcomes represents a major challenge in Crohn's disease (CD). Radiomics provides a method to extract quantitative features from medical images and may successfully predict clinical course. The aim of this pilot study is to evaluate the use of radiomics to predict 10-year surgery for CD patients. We selected a cohort of CD patients with CT scan enterographies and a 10-year follow up. The R library Moddicom was used to extract radiomic features from each lesion of CD, segmented in the CT scans. A logistic regression model based on selected radiomic features was developed to predict 10-year surgery. The model was evaluated by computing the area under the curve (AUC) of the receiver operating characteristic curve, sensitivity, specificity, positive and negative predictive values (PPV, NPV). We enroled 30 patients, with 44 CT scans and 93 lesions. We extracted 217 radiomic features from each lesion. The developed model was based on two radiomic features and presented an AUC (95% CI) of 0.83 (0.73-0.91) in predicting 10-year surgery. Sensitivity, specificity, PPV, NPV of the radiomic model were equal to 0.72, 0.90, 0.79, 0.86, respectively. Radiomics could be a helpful tool to identify patients with high risk for surgery and needing a stricter monitoring.

AI-based pathologic biomarker for pathologic downstaging in patients with muscle-invasive bladder cancer undergoing cystectomy after neoadjuvant nivolumab, gemcitabine, and cisplatin: BLAAST-1 Trial.

e16566 Background: The BLASST-1 study is a multi-center phase II trial evaluating the combination of neoadjuvant nivolumab with gemcitabine-cisplatin (N+GC) for muscle-invasive bladder cancer (MIBC) patients undergoing radical cystectomy (RC). The primary endpoint was pathologic down staging (PaR; ≤pT1N0). We previously reported a PaR rate of 65.8% (Gupta S et al. ASCO GU 2020). Given the lack of validated and optimal biomarkers to predict PaR, we studied the association of an AI-based pathologic biomarker measuring pre-treatment morphological features with PaR. Methods: Forty-one patients with MIBC (cT2-T4a, N≤1, M0) and candidates for RC were enrolled between Feb 2018 and June 2019 (cT2N0 90%, cT3N0 7%, cT4N1 3%). Thirty-six patients had transurethral resection of bladder cancer (TURBT) with pre-treatment diagnostic specimens available for analysis. Patients received four cycles of N+GC followed by RC within 8 weeks. A board-certified pathologist selected diagnostic regions from TURBT-derived representative H&amp;E diagnostic slides for each patient. To compute the pathological biomarker, a proprietary deep-learning algorithm (Valar Labs, Palo Alto, CA) first segmented nuclei from digital whole-slide images of the H&amp;E specimens to extract quantitative histological features. The Valar pathologic biomarker was then computed from features associated with immune infiltration and morphological characteristics of neoplastic cells. The Valar biomarker was split based on unsupervised clustering into two groups: Valar-High was associated with PaR and Valar-Low with no PaR. PD-L1 cutoff of 1% was used for dichotomization. To compare the Valar biomarker with PD-L1, a subcohort of 33 patients with available PD-L1 scores were analyzed. T-tests and diagnostic performance metrics were used to distinguish between PaR response rates in each cluster group across tests. Results: Patients designated Valar-High (n=23) had higher PaR compared to no PaR in the Valar-Low (n=13) group (PAR 78.3% vs 38.4%, respectively, p&lt;0.008). Compared to PD-L1, the Valar biomarker had higher sensitivity (85.7 vs 57.1%), specificity (66.6 vs 66%), positive predictive value (81.8 vs 75.0%), negative predictive value (72.7 vs 47.0%), and accuracy (78.8 vs 60.6%) across both cohorts. The Valar biomarker was not correlated to PD-L1 (r^2=0.029) or sex (22 male, 14 female). The combined Valar-High or PD-L1 High test had a high sensitivity for PaR (95.2%). Conclusions: In this hypothesis-generating study, pre-treatment morphological features measured by the AI-based Valar pathologic biomarker identified responders to neoadjuvant N+GC in MIBC. Further prospective studies are needed to study the prognostic vs. predictive utility of this biomarker.

Radiomics in colorectal cancer patients.

The main therapeutic options for colorectal cancer are surgical resection and adjuvant chemotherapy in non-metastatic disease. However, the evaluation of the overall adjuvant chemotherapy benefit in patients with a high risk of recurrence is challenging. Radiological images can represent a source of data that can be analyzed by using automated computer-based techniques, working on numerical information coded within Digital Imaging and Communications in Medicine files: This image numerical analysis has been named "radiomics". Radiomics allows the extraction of quantitative features from radiological images, mainly invisible to the naked eye, that can be further analyzed by artificial intelligence algorithms. Radiomics is expanding in oncology to either understand tumor biology or for the development of imaging biomarkers for diagnosis, staging, and prognosis, prediction of treatment response and diseases monitoring and surveillance. Several efforts have been made to develop radiomics signatures for colorectal cancer patient using computed tomography (CT) images with different aims: The preoperative prediction of lymph node metastasis, detecting BRAF and RAS gene mutations. Moreover, the use of delta-radiomics allows the analysis of variations of the radiomics parameters extracted from CT scans performed at different timepoints. Most published studies concerning radiomics and magnetic resonance imaging (MRI) mainly focused on the response of advanced tumors that underwent neoadjuvant therapy. Nodes status is the main determinant of adjuvant chemotherapy. Therefore, several radiomics model based on MRI, especially on T2-weighted images and ADC maps, for the preoperative prediction of nodes metastasis in rectal cancer has been developed. Current studies mostly focused on the applications of radiomics in positron emission tomography/CT for the prediction of survival after curative surgical resection and assessment of response following neoadjuvant chemoradiotherapy. Since colorectal liver metastases develop in about 25% of patients with colorectal carcinoma, the main diagnostic tasks of radiomics should be the detection of synchronous and metachronous lesions. Radiomics could be an additional tool in clinical setting, especially in identifying patients with high-risk disease. Nevertheless, radiomics has numerous shortcomings that make daily use extremely difficult. Further studies are needed to assess performance of radiomics in stratifying patients with high-risk disease.

Modeling the regulation of the Il12b locus by NF-κB dimers

Abstract The transcription factor NF-κB is composed of a family, which regulates the expression of various genes in immune and inflammatory responses. The NF-κB family consists of five members: RelA/p65, RelB, c-Rel, p50/p105, and p52/p100, which control the transcription of target genes by binding to regulatory κB sites in homo-or heterodimer forms. Unlike the widely known RelA, the mechanisms related to the regulation of target genes by c-Rel is less understood. Interleukin 12b(Il12b), which encodes the p40 subunit of IL12 and IL23, is one example of gene that depend on c-Rel for proper activation in macrophages. The differences in intrinsic binding properties of c-Rel- and RelA-containing dimers at various regulatory sites around Il12b can be a key to understand the mechanism of c-Rel-containing dimers’ role in Il12b regulation. Using c-Rel and Il12b reporter mice with fluorescence live-cell imaging, we observe the single-cell dynamics of nuclear c-Rel and Il12b gene expression in bone marrow-derived macrophages induced by NF-κB activating ligands. We analyze the live-cell imaging data and extract quantitative features of nuclear c-Rel and Il12b expression from the time-series data. For additional mechanistic insights of the Il12b gene regulation by NF-κB dimers, we propose a mathematical model of dynamic gene regulation which includes actions of RelA- and c-Rel-containing dimers at κB sites proximal and distal to the Il12b locus. Our combined experimental and mathematical modeling approach will reveal quantitative insights that may provide a more comprehensive understanding of regulatory mechanisms underlying Il12b gene expression employed by NF-κB dimers.

Abstract 5783: Integration of AI-powered digital pathology and imaging mass cytometry to identify relevant features of the tumor microenvironment

Abstract Digital pathology coupled to artificial intelligence (AI)-powered approaches are receiving great attention in the oncoimmunology field, as their adoption holds promise to improve current diagnostic workflows and potentiate the analytic outputs. In this work, we aimed at combining different histopathological approaches and AI-aided analytic tools to analyze the ecosystem of tumor tissues. By deploying AI-powered standard H&amp;E and high-dimensional imaging-mass cytometry (IMC) to FFPE tissue samples, we could extract quantitative and standardized features that couldn’t have been easily identified and integrated by eye. One tissue microarray (TMA) slide containing 108 spots of NSCLC specimens (both adenocarcinoma and squamous carcinoma) was stained with H&amp;E and scanned through the Axio Scan.Z1 (ZEISS) to generate high-quality virtual images. A deep learning algorithm was trained and applied to H&amp;E images to identify tumor cells. The consecutive tissue section was stained with metal-labeled antibodies and processed through the Hyperion workflow (StandardBiotools), allowing quantitative detection of a panel of 23 markers related to tumor cells (Pan-cytokeratin), tissue architecture (aSMA, Vimentin, CD31, Collagen I, nuclei), CD45+ immune cells, comprehensive of myeloid cells (CD68, CD14, CD16, CD163, CD63, CCR4), lymphoid cells (CD3, CD4, CD8, FOXP3, CD20) and immune activation (S100A8, HLA-DR, Granzyme-B, KI67, Arginase-1). Data were exported as MCD files, visualized using the MCD viewer and further analyzed with the Qupath software. Cell segmentation was performed by the CellProfiler and Ilastik softwares and main cell populations were identified by a supervised approach through Cytomap. On H&amp;E images, we generated a classifier of tumor heterogeneity, by exploring the spatial localization of tumor cells with the K-function summary statistic, which analyzes the distribution of tumor cells as a function of their distance. The resulting K-score value was then used to classify each tumor spot as diffuse, poorly clustered or highly clustered. Multiparametric computational analysis of the IMC images allowed to grasp immune and stromal classifiers, including frequency of immune cell populations in the tumor nests versus fibrotic stroma and immune cell interactions. In conclusion, AI-powered analysis of H&amp;E slides is a robust approach that can improve manual scoring and unlock tissue relevant features opening to new diagnostic possibilities. Meanwhile, the analysis of the immune ecosystem by multiparametric imaging mass cytometry allows investigating spatial patterns and cell interactions at single-cell level. Integration of these approaches is feasible and allows the identification of tumor patient profiles with clinical relevance. Citation Format: Alessandra Rigamonti, Marika Viatore, Rebecca Polidori, Marco Erreni, Maria Fumagalli, Daoud Rahal, Massimo Locati, Alberto Mantovani, Federica Marchesi. Integration of AI-powered digital pathology and imaging mass cytometry to identify relevant features of the tumor microenvironment. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5783.

Deep learning-powered 3D segmentation derives factors associated with lymphovascular invasion and prognosis in clinical T1 stage non-small cell lung cancer

BackgroundLymphovascular invasion (LVI) is an invasive biologic behavior that affects the treatment and prognosis of patients with early-stage lung cancer. This study aimed to identify LVI diagnostic and prognostic biomarkers using deep learning-powered 3D segmentation with artificial intelligence (AI) technology. MethodsBetween January 2016 and October 2021, we enrolled patients with clinical T1 stage non-small cell lung cancer (NSCLC). We used commercially available AI software (Dr. Wise system, Deep-wise Corporation, China) to extract quantitative AI features of pulmonary nodules automatically. Dimensionality reduction was achieved through least absolute shrinkage and selection operator regression; subsequently, the AI score was calculated.Then, the univariate and multivariate analysis was further performed on the AI score and patient baseline parameters. ResultsAmong 175 enrolled patients, 22 tested positive for LVI at pathology review. Based on the multivariate logistic regression results, we incorporated the AI score, carcinoembryonic antigen, spiculation, and pleural indentation into the nomogram for predicting LVI. The nomogram showed good discrimination (C-index = 0.915 [95% confidence interval: 0.89–0.94]); moreover, calibration of the nomogram revealed good predictive ability (Brier score = 0.072). Kaplan–Meier analysis revealed that relapse-free survival and overall survival were significantly higher among patients with a low-risk AI score and without LVI than those among patients with a high-risk AI score (p = 0.008 and p = 0.002, respectively) and with LVI (p = 0.013 and p = 0.008, respectively). ConclusionsOur findings indicate that a high-risk AI score is a diagnostic biomarker for LVI in patients with clinical T1 stage NSCLC; accordingly, it can serve as a prognostic biomarker for these patients.

Radiomics in Hypopharyngeal Cancer Management: A State-of-the-Art Review.

(1) Background: Hypopharyngeal squamous cell carcinomas usually present with locally advanced disease and a correspondingly poor prognosis. Currently, efforts are being made to improve tumor characterization and provide insightful information for outcome prediction. Radiomics is an emerging area of study that involves the conversion of medical images into mineable data; these data are then used to extract quantitative features based on shape, intensity, texture, and other parameters; (2) Methods: A systematic review of the peer-reviewed literature was conducted; (3) Results: A total of 437 manuscripts were identified. Fifteen manuscripts met the inclusion criteria. The main targets described were the evaluation of textural features to determine tumor-programmed death-ligand 1 expression; a surrogate for microvessel density and heterogeneity of perfusion; patient stratification into groups at high and low risk of progression; prediction of early recurrence, 1-year locoregional failure and survival outcome, including progression-free survival and overall survival, in patients with locally advanced HPSCC; thyroid cartilage invasion, early disease progression, recurrence, induction chemotherapy response, treatment response, and prognosis; and (4) Conclusions: our findings suggest that radiomics represents a potentially useful tool in the diagnostic workup as well as during the treatment and follow-up of patients with HPSCC. Large prospective studies are essential to validate this technology in these patients.

Abstract P5-02-35: Radiomic biomarkers to predict the efficacy of anti-PD-1 immunotherapy-based combinational treatment in advanced breast cancer: a multi-center study

Abstract Introduction: Immunotherapy, especially immune checkpoint inhibitors, is regarded as one of the major breakthroughs in breast cancer treatment. However, it is an important challenge to accurately locate the patients who benefit from immunotherapy, because there is still a lack of universal and robust predictors of the efficacy of immunotherapy. Radiomics can extract quantitative imaging features in a highthroughput manner and assess tumor microenvironment and heterogeneity. This study investigated the correlation between deep learning radiomic biomarkers, including its predictive value for immunotherapy response in advanced breast cancer (ABC) patients. Methods: 240 patients with metastatic breast cancer treated with anti-PD-1 immunotherapy in three institutions from February 2018 to January 2022 were studied retrospectively, among which, the data of 61 patients were collected through prospective clinical trials. For these data, 189 ABC patients from prospective clinical trials and Sun Yat-sen University Cancer Center were evaluated as a training set to establish a radiomic model to predict value of immunotherapy, then this model was independently validated with 51 ABC patients from Sun Yat-sen Memorial Hospital. The CE-CT (contrast enhanced computed tomography) images of patients within one month before immunotherapy were were delineated with regions of interest (ROI) and radiomics features extraction. Data dimension reduction, feature selection and radiomic model construction were carried out with multilayer perceptron (MLP) deep learning. Combined with the radiomics signatures, independent clinical characteristics and pathological risk factors, the predictive model was established by multivariable logistic regression analysis. ROC curve (receiver operator area under receiver operator area, AUC) and Delong test were used to evaluate and compare the prediction performance of the model. Finally, decision curve analysis (DCA) is used to determine the net benefits predicted by the model. Results: The radiomic biomarker performed well in predicting response to immunotherapy, reflflected by the AUCs in the training set(AUC=0.885, 95% CI: 0.829-0.941) and validation set (AUC=0.871, 95% CI: 0.752-0.991), respectively. The accuracy of this radiomics model was better than those of clinical indicators, including PD-L1 expression. Conclusions: By combining deep learning technology and CT images and PD-L1 expression, we developed an independent predictive model that could identify MBC patients most likely to benefifit from immunotherapy, and may effectively improve more precise and individualized decision support. Citation Format: Jieqiong Liu, Jianli Zhao, Zhixian Sun, Yunfang Yu, Zhongyu Yuan, Herui Yao, Ying Wang. Radiomic biomarkers to predict the efficacy of anti-PD-1 immunotherapy-based combinational treatment in advanced breast cancer: a multi-center study [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P5-02-35.

Radiomics-guided checkpoint inhibitor immunotherapy for precision medicine in cancer: A review for clinicians.

Immunotherapy using immune checkpoint inhibitors (ICIs) is a breakthrough in oncology development and has been applied to multiple solid tumors. However, unlike traditional cancer treatment approaches, immune checkpoint inhibitors (ICIs) initiate indirect cytotoxicity by generating inflammation, which causes enlargement of the lesion in some cases. Therefore, rather than declaring progressive disease (PD) immediately, confirmation upon follow-up radiological evaluation after four-eight weeks is suggested according to immune-related Response Evaluation Criteria in Solid Tumors (ir-RECIST). Given the difficulty for clinicians to immediately distinguish pseudoprogression from true disease progression, we need novel tools to assist in this field. Radiomics, an innovative data analysis technique that quantifies tumor characteristics through high-throughput extraction of quantitative features from images, can enable the detection of additional information from early imaging. This review will summarize the recent advances in radiomics concerning immunotherapy. Notably, we will discuss the potential of applying radiomics to differentiate pseudoprogression from PD to avoid condition exacerbation during confirmatory periods. We also review the applications of radiomics in hyperprogression, immune-related biomarkers, efficacy, and immune-related adverse events (irAEs). We found that radiomics has shown promising results in precision cancer immunotherapy with early detection in noninvasive ways.

Child-therapist interaction features impact Autism treatment response trajectories

BackgroundIdentifying mechanisms of change in Autism treatment may help explain response variability and maximize efficacy. For this, the child-therapist interaction could have a key role as stressed by developmental models of intervention, but still remains under-investigated. AimsThe longitudinal study of treatment response trajectories considering both baseline and child-therapist interaction features by means of predictive modeling. Methods and ProceduresN = 25 preschool children were monitored for one year during Naturalistic Developmental Behavioral Intervention. N = 100 video-recorded sessions were annotated with an observational coding system at four time points, to extract quantitative interaction features. Outcomes and ResultsBaseline and interaction variables were combined to predict response trajectories at one year, and achieved the best predictive performance. The baseline developmental gap, therapist’s efficacy in child engagement, respecting children’s timing after fast behavioral synchronization, and modulating the interplay to prevent child withdrawal emerged as key factors. Further, changes in interaction patterns in the early phase of the intervention were predictive of the overall response to treatment. Conclusions and ImplicationsClinical implications are discussed, stressing the importance of promoting emotional self-regulation during intervention and the possible relevance of the first period of intervention for later response.