RESUMO

OBJECTIVES: To validate the performance of Mirai, a mammography-based deep learning model, in predicting breast cancer risk over a 1-5-year period in Mexican women. METHODS: This retrospective single-center study included mammograms in Mexican women who underwent screening mammography between January 2014 and December 2016. For women with consecutive mammograms during the study period, only the initial mammogram was included. Pathology and imaging follow-up served as the reference standard. Model performance in the entire dataset was evaluated, including the concordance index (C-Index) and area under the receiver operating characteristic curve (AUC). Mirai's performance in terms of AUC was also evaluated between mammography systems (Hologic versus IMS). Clinical utility was evaluated by determining a cutoff point for Mirai's continuous risk index based on identifying the top 10% of patients in the high-risk category. RESULTS: Of 3110 patients (median age 52.6 years ± 8.9), throughout the 5-year follow-up period, 3034 patients remained cancer-free, while 76 patients developed breast cancer. Mirai achieved a C-index of 0.63 (95% CI: 0.6-0.7) for the entire dataset. Mirai achieved a higher mean C-index in the Hologic subgroup (0.63 [95% CI: 0.5-0.7]) versus the IMS subgroup (0.55 [95% CI: 0.4-0.7]). With a Mirai index score > 0.029 (10% threshold) to identify high-risk individuals, the study revealed that individuals in the high-risk group had nearly three times the risk of developing breast cancer compared to those in the low-risk group. CONCLUSIONS: Mirai has a moderate performance in predicting future breast cancer among Mexican women. CRITICAL RELEVANCE STATEMENT: Prospective efforts should refine and apply the Mirai model, especially to minority populations and women aged between 30 and 40 years who are currently not targeted for routine screening. KEY POINTS: The applicability of AI models to non-White, minority populations remains understudied. The Mirai model is linked to future cancer events in Mexican women. Further research is needed to enhance model performance and establish usage guidelines.

RESUMO

Introduction: Obesity and gender play a critical role in shaping the outcomes of COVID-19 disease. These two factors have a dynamic relationship with each other, as well as other risk factors, which hinders interpretation of how they influence severity and disease progression. This work aimed to study differences in COVID-19 disease outcomes through analysis of risk profiles stratified by gender and obesity status. Methods: This study employed an unsupervised clustering analysis, using Mexico's national COVID-19 hospitalization dataset, which contains demographic information and health outcomes of patients hospitalized due to COVID-19. Patients were segmented into four groups by obesity and gender, with participants' attributes and clinical outcome data described for each. Then, Consensus and PAM clustering methods were used to identify distinct risk profiles based on underlying patient characteristics. Risk profile discovery was completed on 70% of records, with the remaining 30% available for validation. Results: Data from 88,536 hospitalized patients were analyzed. Obesity, regardless of gender, was linked with higher odds of hypertension, diabetes, cardiovascular diseases, pneumonia, and Intensive Care Unit (ICU) admissions. Men tended to have higher frequencies of ICU admissions and pneumonia and higher mortality rates than women. Within each of the four analysis groups (divided based on gender and obesity status), clustering analyses identified four to five distinct risk profiles. For example, among women with obesity, there were four profiles; those with a hypertensive profile were more likely to have pneumonia, and those with a diabetic profile were most likely to be admitted to the ICU. Conclusion: Our analysis emphasizes the complex interplay between obesity, gender, and health outcomes in COVID-19 hospitalizations. The identified risk profiles highlight the need for personalized treatment strategies for COVID-19 patients and can assist in planning for patterns of deterioration in future waves of SARS-CoV-2 virus transmission. This research underscores the importance of tackling obesity as a major public health concern, given its interplay with many other health conditions, including infectious diseases such as COVID-19.

Assuntos

RESUMO

Background: Breast cancer is the leading cause of cancer-related fatalities among women worldwide. Conventional screening and risk prediction models primarily rely on demographic and patient clinical history to devise policies and estimate likelihood. However, recent advancements in artificial intelligence (AI) techniques, particularly deep learning (DL), have shown promise in the development of personalized risk models. These models leverage individual patient information obtained from medical imaging and associated reports. In this systematic review, we thoroughly investigated the existing literature on the application of DL to digital mammography, radiomics, genomics, and clinical information for breast cancer risk assessment. We critically analyzed these studies and discussed their findings, highlighting the promising prospects of DL techniques for breast cancer risk prediction. Additionally, we explored ongoing research initiatives and potential future applications of AI-driven approaches to further improve breast cancer risk prediction, thereby facilitating more effective screening and personalized risk management strategies. Objective and methods: This study presents a comprehensive overview of imaging and non-imaging features used in breast cancer risk prediction using traditional and AI models. The features reviewed in this study included imaging, radiomics, genomics, and clinical features. Furthermore, this survey systematically presented DL methods developed for breast cancer risk prediction, aiming to be useful for both beginners and advanced-level researchers. Results: A total of 600 articles were identified, 20 of which met the set criteria and were selected. Parallel benchmarking of DL models, along with natural language processing (NLP) applied to imaging and non-imaging features, could allow clinicians and researchers to gain greater awareness as they consider the clinical deployment or development of new models. This review provides a comprehensive guide for understanding the current status of breast cancer risk assessment using AI. Conclusion: This study offers investigators a different perspective on the use of AI for breast cancer risk prediction, incorporating numerous imaging and non-imaging features.

RESUMO

Training deep Convolutional Neural Networks (CNNs) presents challenges in terms of memory requirements and computational resources, often resulting in issues such as model overfitting and lack of generalization. These challenges can only be mitigated by using an excessive number of training images. However, medical image datasets commonly suffer from data scarcity due to the complexities involved in their acquisition, preparation, and curation. To address this issue, we propose a compact and hybrid machine learning architecture based on the Morphological and Convolutional Neural Network (MCNN), followed by a Random Forest classifier. Unlike deep CNN architectures, the MCNN was specifically designed to achieve effective performance with medical image datasets limited to a few hundred samples. It incorporates various morphological operations into a single layer and uses independent neural networks to extract information from each signal channel. The final classification is obtained by utilizing a Random Forest classifier on the outputs of the last neural network layer. We compare the classification performance of our proposed method with three popular deep CNN architectures (ResNet-18, ShuffleNet-V2, and MobileNet-V2) using two training approaches: full training and transfer learning. The evaluation was conducted on two distinct medical image datasets: the ISIC dataset for melanoma classification and the ORIGA dataset for glaucoma classification. Results demonstrate that the MCNN method exhibits reliable performance in melanoma classification, achieving an AUC of 0.94 (95% CI: 0.91 to 0.97), outperforming the popular CNN architectures. For the glaucoma dataset, the MCNN achieved an AUC of 0.65 (95% CI: 0.53 to 0.74), which was similar to the performance of the popular CNN architectures. This study contributes to the understanding of mathematical morphology in shallow neural networks for medical image classification and highlights the potential of hybrid architectures in effectively learning from medical image datasets that are limited by a small number of case samples.

RESUMO

BACKGROUND: Recent advancements in computing power and state-of-the-art algorithms have helped in more accessible and accurate diagnosis of numerous diseases. In addition, the development of de novo areas in imaging science, such as radiomics and radiogenomics, have been adding more to personalize healthcare to stratify patients better. These techniques associate imaging phenotypes with the related disease genes. Various imaging modalities have been used for years to diagnose breast cancer. Nonetheless, digital breast tomosynthesis (DBT), a state-of-the-art technique, has produced promising results comparatively. DBT, a 3D mammography, is replacing conventional 2D mammography rapidly. This technological advancement is key to AI algorithms for accurately interpreting medical images. OBJECTIVE AND METHODS: This paper presents a comprehensive review of deep learning (DL), radiomics and radiogenomics in breast image analysis. This review focuses on DBT, its extracted synthetic mammography (SM), and full-field digital mammography (FFDM). Furthermore, this survey provides systematic knowledge about DL, radiomics, and radiogenomics for beginners and advanced-level researchers. RESULTS: A total of 500 articles were identified, with 30 studies included as the set criteria. Parallel benchmarking of radiomics, radiogenomics, and DL models applied to the DBT images could allow clinicians and researchers alike to have greater awareness as they consider clinical deployment or development of new models. This review provides a comprehensive guide to understanding the current state of early breast cancer detection using DBT images. CONCLUSION: Using this survey, investigators with various backgrounds can easily seek interdisciplinary science and new DL, radiomics, and radiogenomics directions towards DBT.

Assuntos

RESUMO

PURPOSE: Multiomics cancer subtyping is becoming increasingly popular for directing state-of-the-art therapeutics. However, these methods have never been systematically assessed for their ability to capture cancer prognosis for identified subtypes, which is essential to effectively treat patients. METHODS: We systematically searched PubMed, The Cancer Genome Atlas, and Pan-Cancer Atlas for multiomics cancer subtyping studies from 2010 through 2019. Studies comprising at least 50 patients and examining survival were included. Pooled Cox and logistic mixed-effects models were used to compare the ability of multiomics subtyping methods to identify clinically prognostic subtypes, and a structural equation model was used to examine causal paths underlying subtyping method and mortality. RESULTS: A total of 31 studies comprising 10,848 unique patients across 32 cancers were analyzed. Latent-variable subtyping was significantly associated with overall survival (adjusted hazard ratio, 2.81; 95% CI, 1.16-6.83; P = .023) and vital status (1 year adjusted odds ratio, 4.71; 95% CI, 1.34-16.49; P = .015; 5 year adjusted odds ratio, 7.69; 95% CI, 1.83-32.29; P = .005); latent-variable-identified subtypes had greater associations with mortality across models (adjusted hazard ratio, 1.19; 95% CI, 1.01-1.42; P = .050). Our structural equation model confirmed the path from subtyping method through multiomics subtype (߈ = 0.66; P = .048) on survival (߈ = 0.37; P = .008). CONCLUSION: Multiomics methods have different abilities to define clinically prognostic cancer subtypes, which should be considered before administration of personalized therapy; preliminary evidence suggests that latent-variable methods better identify clinically prognostic biomarkers and subtypes.

Assuntos

RESUMO

BACKGROUND: Alzheimer's Disease (AD) is an irreversible, progressive brain disorder that slowly destroys memory and thinking skills. The ability to correctly predict the diagnosis of Alzheimer's disease in its earliest stages can help physicians make more informed clinical decisions on therapy plans. OBJECTIVE: This study aimed to determine whether the unsupervised discovering of latent classes of subjects with Mild Cognitive Impairment (MCI) may be useful in finding different prodromal AD stages and/or subjects with a low MCI to AD conversion risk. METHODS: Total 18 features relevant to the MCI to AD conversion process led to the identification of 681 subjects with early MCI. Subjects were divided into training (70%) and validation (30%) sets. Subjects from the training set were analyzed using consensus clustering, and Gaussian Mixture Models (GMM) were used to describe the latent classes. The discovered GMM predicted the latent class of the validation set. Finally, descriptive statistics, rates of conversion, and Odds Ratios (OR) were computed for each discovered class. RESULTS: Through consensus clustering, we discovered three different clusters among MCI subjects. The three clusters were associated with low-risk (OR = 0.12, 95%CI = 0.04 to 0.3|), medium-risk (OR = 1.33, 95%CI = 0.75 to 2.37), and high-risk (OR = 3.02, 95%CI = 1.64 to 5.57) of converting from MCI to AD, with the high-risk and low-risk groups highly contrasting. Hence, prodromal AD subjects were present in only two clusters. CONCLUSION: We successfully discovered three different latent classes among MCI subjects with varied risks of MCI-to-AD conversion through consensus clustering. Two of the discovered classes may represent two different prodromal presentations of Alzheimer´s disease.

Assuntos

RESUMO

Purpose: To determine and characterize the radiomics features from structural MRI (MPRAGE) and Diffusion Tensor Imaging (DTI) associated with the presence of mild traumatic brain injuries on student athletes with post-concussive syndrome (PCS). Material and Methods: 122 student athletes (65 M, 57 F), median (IQR) age 18.8 (15-20) years, with a mixed level of play and sports activities, with a known history of concussion and clinical PCS, and 27 (15 M, 12 F), median (IQR) age 20 (19, 21) years, concussion free athlete subjects were MRI imaged in a clinical MR machine. MPRAGE and DTI-FA and DTI-ADC images were used to extract radiomic features from white and gray matter regions within the entire brain (2 ROI) and the eight main lobes of the brain (16 ROI) for a total of 18 analyzed regions. Radiomic features were divided into five different data sets used to train and cross-validate five different filter-based Support Vector Machines. The top selected features of the top model were described. Furthermore, the test predictions of the top four models were ensembled into a single average prediction. The average prediction was evaluated for the association to the number of concussions and time from injury. Results: Ninety-one PCS subjects passed inclusion criteria (91 Cases, 27 controls). The average prediction of the top four models had a sensitivity of 0.80, 95% CI: [0.71, 0.88] and specificity of 0.74 95%CI [0.54, 0.89] for distinguishing subjects from controls. The white matter features were strongly associated with mTBI, while the whole-brain analysis of gray matter showed the worst association. The predictive index was significantly associated with the number of concussions (p < 0.0001) and associated with the time from injury (p < 0.01). Conclusion: MRI Radiomic features are associated with a history of mTBI and they were successfully used to build a predictive machine learning model for mTBI for subjects with PCS associated with a history of one or more concussions.

RESUMO

Late-onset Alzheimer's Disease (LOAD) is the most common form of dementia in the elderly. Genome-wide association studies (GWAS) for LOAD have open new avenues to identify genetic causes and to provide diagnostic tools for early detection. Although several predictive models have been proposed using the few detected GWAS markers, there is still a need for improvement and identification of potential markers. Commonly, polygenic risk scores are being used for prediction. Nevertheless, other methods to generate predictive models have been suggested. In this research, we compared three machine learning methods that have been proved to construct powerful predictive models (genetic algorithms, LASSO, and step-wise) and propose the inclusion of markers from misclassified samples to improve overall prediction accuracy. Our results show that the addition of markers from an initial model plus the markers of the model fitted to misclassified samples improves the area under the receiving operative curve by around 5%, reaching ~0.84, which is highly competitive using only genetic information. The computational strategy used here can help to devise better methods to improve classification models for AD. Our results could have a positive impact on the early diagnosis of Alzheimer's disease.

Assuntos

RESUMO

BACKGROUND: Late-Onset Alzheimer's Disease (LOAD) is a leading form of dementia. There is no effective cure for LOAD, leaving the treatment efforts to depend on preventive cognitive therapies, which stand to benefit from the timely estimation of the risk of developing the disease. Fortunately, a growing number of Machine Learning methods that are well positioned to address this challenge are becoming available. RESULTS: We conducted systematic comparisons of representative Machine Learning models for predicting LOAD from genetic variation data provided by the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort. Our experimental results demonstrate that the classification performance of the best models tested yielded ∼72% of area under the ROC curve. CONCLUSIONS: Machine learning models are promising alternatives for estimating the genetic risk of LOAD. Systematic machine learning model selection also provides the opportunity to identify new genetic markers potentially associated with the disease.

Assuntos

RESUMO

BACKGROUND: Diagnosing Alzheimer's disease (AD) in its earliest stages is important for therapeutic and support planning. Similarly, being able to predict who will convert from mild cognitive impairment (MCI) to AD would have clinical implications. OBJECTIVES: The goals of this study were to identify features from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database associated with the conversion from MCI to AD, and to characterize the temporal evolution of that conversion. METHODS: We screened the publically available ADNI longitudinal database for subjects with MCI who have developed AD (cases: n=305), and subjects with MCI who have remained stable (controls: n=250). Analyses included 1,827 features from laboratory assays (n=12), quantitative MRI scans (n=1,423), PET studies (n=136), medical histories (n=72), and neuropsychological tests (n=184). Statistical longitudinal models identified features with significant differences in longitudinal behavior between cases and matched controls. A multiple-comparison adjusted log-rank test identified the capacity of the significant predictive features to predict early conversion. RESULTS: 411 features (22.5%) were found to be statistically different between cases and controls at the time of AD diagnosis; 385 features were statistically different at least 6 months prior to diagnosis, and 28 features distinguished early from late conversion, 20 of which were obtained from neuropsychological tests. In addition, 69 features (3.7%) had statistically significant changes prior to AD diagnosis. CONCLUSION: Our results characterized features associated with disease progression from MCI to AD, and, in addition, the log-rank test identified features which are associated with the risk of early conversion.

Assuntos

RESUMO

Previous methods proposed for the detection of cancer driver mutations have been based on the estimation of background mutation rate, impact on protein function, or network influence. In this paper, we instead focus on those factors influencing patient survival. To this end, an approximation of the log-rank test has been systematically applied, even though it assumes a large and similar number of patients in both risk groups, which is violated in cancer genomics. Here, we propose VALORATE, a novel algorithm for the estimation of the null distribution for the log-rank, independent of the number of mutations. VALORATE is based on conditional distributions of the co-occurrences between events and mutations. The results, achieved through simulations, comparisons with other methods, analyses of TCGA and ICGC cancer datasets, and validations, suggest that VALORATE is accurate, fast, and can identify both known and novel gene mutations. Our proposal and results may have important implications in cancer biology, bioinformatics analyses, and ultimately precision medicine.

Assuntos

RESUMO

In this work, the potential of X-ray based multivariate prognostic models to predict the onset of chronic knee pain is presented. Using X-rays quantitative image assessments of joint-space-width (JSW) and paired semiquantitative central X-ray scores from the Osteoarthritis Initiative (OAI), a case-control study is presented. The pain assessments of the right knee at the baseline and the 60-month visits were used to screen for case/control subjects. Scores were analyzed at the time of pain incidence (T-0), the year prior incidence (T-1), and two years before pain incidence (T-2). Multivariate models were created by a cross validated elastic-net regularized generalized linear models feature selection tool. Univariate differences between cases and controls were reported by AUC, C-statistics, and ODDs ratios. Univariate analysis indicated that the medial osteophytes were significantly more prevalent in cases than controls: C-stat 0.62, 0.62, and 0.61, at T-0, T-1, and T-2, respectively. The multivariate JSW models significantly predicted pain: AUC = 0.695, 0.623, and 0.620, at T-0, T-1, and T-2, respectively. Semiquantitative multivariate models predicted paint with C-stat = 0.671, 0.648, and 0.645 at T-0, T-1, and T-2, respectively. Multivariate models derived from plain X-ray radiography assessments may be used to predict subjects that are at risk of developing knee pain.

Assuntos

RESUMO

Mammography is the most common and effective breast cancer screening test. However, the rate of positive findings is very low, making the radiologic interpretation monotonous and biased toward errors. This work presents a computer-aided diagnosis (CADx) method aimed to automatically triage mammogram sets. The method coregisters the left and right mammograms, extracts image features, and classifies the subjects into risk of having malignant calcifications (CS), malignant masses (MS), and healthy subject (HS). In this study, 449 subjects (197 CS, 207 MS, and 45 HS) from a public database were used to train and evaluate the CADx. Percentile-rank (p-rank) and z-normalizations were used. For the p-rank, the CS versus HS model achieved a cross-validation accuracy of 0.797 with an area under the receiver operating characteristic curve (AUC) of 0.882; the MS versus HS model obtained an accuracy of 0.772 and an AUC of 0.842. For the z-normalization, the CS versus HS model achieved an accuracy of 0.825 with an AUC of 0.882 and the MS versus HS model obtained an accuracy of 0.698 and an AUC of 0.807. The proposed method has the potential to rank cases with high probability of malignant findings aiding in the prioritization of radiologists work list.

Assuntos

RESUMO

The early diagnosis of Alzheimer's disease (AD) and mild cognitive impairment (MCI) is very important for treatment research and patient care purposes. Few biomarkers are currently considered in clinical settings, and their use is still optional. The objective of this work was to determine whether multimodal and nonpreviously AD associated features could improve the classification accuracy between AD, MCI, and healthy controls, which may impact future AD biomarkers. For this, Alzheimer's Disease Neuroimaging Initiative database was mined for case-control candidates. At least 652 baseline features extracted from MRI and PET analyses, biological samples, and clinical data up to February 2014 were used. A feature selection methodology that includes a genetic algorithm search coupled to a logistic regression classifier and forward and backward selection strategies was used to explore combinations of features. This generated diagnostic models with sizes ranging from 3 to 8, including well documented AD biomarkers, as well as unexplored image, biochemical, and clinical features. Accuracies of 0.85, 0.79, and 0.80 were achieved for HC-AD, HC-MCI, and MCI-AD classifications, respectively, when evaluated using a blind test set. In conclusion, a set of features provided additional and independent information to well-established AD biomarkers, aiding in the classification of MCI and AD.

Assuntos

RESUMO

One of the initial and critical procedures for the analysis of metabolomics data using liquid chromatography and mass spectrometry is feature detection. Feature detection is the process to detect boundaries of the mass surface from raw data. It consists of detected abundances arranged in a two-dimensional (2D) matrix of mass/charge and elution time. MZmine 2 is one of the leading software environments that provide a full analysis pipeline for these data. However, the feature detection algorithms provided in MZmine 2 are based mainly on the analysis of one-dimension at a time. We propose GridMass, an efficient algorithm for 2D feature detection. The algorithm is based on landing probes across the chromatographic space that are moved to find local maxima providing accurate boundary estimations. We tested GridMass on a controlled marker experiment, on plasma samples, on plant fruits, and in a proteome sample. Compared with other algorithms, GridMass is faster and may achieve comparable or better sensitivity and specificity. As a proof of concept, GridMass has been implemented in Java under the MZmine 2 environment and is available at http://www.bioinformatica.mty.itesm.mx/GridMass and MASSyPup. It has also been submitted to the MZmine 2 developing community.

Assuntos

RESUMO

Early diagnoses of Alzheimer's disease (AD) would confer many benefits. Several biomarkers have been proposed to achieve such a task, where features extracted from magnetic resonance imaging (MRI) have played an important role. However, studies have focused exclusively on morphological characteristics. This study aims to determine whether features relating to the signal and texture of the image could predict mild cognitive impairment (MCI) to AD progression. Clinical, biological, and positron emission tomography information and MRI images of 62 subjects from the AD neuroimaging initiative were used in this study, extracting 4150 features from each MRI. Within this multimodal database, a feature selection algorithm was used to obtain an accurate and small logistic regression model, generated by a methodology that yielded a mean blind test accuracy of 0.79. This model included six features, five of them obtained from the MRI images, and one obtained from genotyping. A risk analysis divided the subjects into low-risk and high-risk groups according to a prognostic index. The groups were statistically different ([Formula: see text]). These results demonstrated that MRI features related to both signal and texture add MCI to AD predictive power, and supported the ongoing notion that multimodal biomarkers outperform single-modality ones.