RESUMEN
PURPOSE: Medical image analysis has become a prominent area where machine learning has been applied. However, high-quality, publicly available data are limited either due to patient privacy laws or the time and cost required for experts to annotate images. In this retrospective study, we designed and evaluated a pipeline to generate synthetic labeled polyp images for augmenting medical image segmentation models with the aim of reducing this data scarcity. METHODS: We trained diffusion models on the HyperKvasir dataset, comprising 1000 images of polyps in the human GI tract from 2008 to 2016. Qualitative expert review, Fréchet Inception Distance (FID), and Multi-Scale Structural Similarity (MS-SSIM) were tested for evaluation. Additionally, various segmentation models were trained with the generated data and evaluated using Dice score (DS) and Intersection over Union (IoU). RESULTS: Our pipeline produced images more akin to real polyp images based on FID scores. Segmentation model performance also showed improvements over GAN methods when trained entirely, or partially, with synthetic data, despite requiring less compute for training. Moreover, the improvement persists when tested on different datasets, showcasing the transferability of the generated images. CONCLUSIONS: The proposed pipeline produced realistic image and mask pairs which could reduce the need for manual data annotation when performing a machine learning task. We support this use case by showing that the methods proposed in this study enhanced segmentation model performance, as measured by Dice and IoU scores, when trained fully or partially on synthetic data.
Asunto(s)
Aprendizaje Automático , Humanos , Estudios Retrospectivos , Procesamiento de Imagen Asistido por Computador/métodosRESUMEN
In recent years, Artificial Intelligence has been used to assist healthcare professionals in detecting and diagnosing neurodegenerative diseases. In this study, we propose a methodology to analyze functional Magnetic Resonance Imaging signals and perform classification between Parkinson's disease patients and healthy participants using Machine Learning algorithms. In addition, the proposed approach provides insights into the brain regions affected by the disease. The functional Magnetic Resonance Imaging from the PPMI and 1000-FCP datasets were pre-processed to extract time series from 200 brain regions per participant, resulting in 11,600 features. Causal Forest and Wrapper Feature Subset Selection algorithms were used for dimensionality reduction, resulting in a subset of features based on their heterogeneity and association with the disease. We utilized Logistic Regression and XGBoost algorithms to perform PD detection, achieving 97.6% accuracy, 97.5% F1 score, 97.9% precision, and 97.7%recall by analyzing sets with fewer than 300 features in a population including men and women. Finally, Multiple Correspondence Analysis was employed to visualize the relationships between brain regions and each group (women with Parkinson, female controls, men with Parkinson, male controls). Associations between the Unified Parkinson's Disease Rating Scale questionnaire results and affected brain regions in different groups were also obtained to show another use case of the methodology. This work proposes a methodology to (1) classify patients and controls with Machine Learning and Causal Forest algorithm and (2) visualize associations between brain regions and groups, providing high-accuracy classification and enhanced interpretability of the correlation between specific brain regions and the disease across different groups.
Asunto(s)
Aprendizaje Automático , Imagen por Resonancia Magnética , Enfermedad de Parkinson , Humanos , Enfermedad de Parkinson/diagnóstico por imagen , Enfermedad de Parkinson/fisiopatología , Imagen por Resonancia Magnética/métodos , Masculino , Femenino , Persona de Mediana Edad , Anciano , Algoritmos , Encéfalo/diagnóstico por imagen , Encéfalo/fisiopatologíaRESUMEN
Background: Electrocardiographic early repolarization (EER) is linked with idiopathic ventricular fibrillation in adults. It is frequently seen in children, with poorly understood significance. Some evidence suggests that it could be a vagally mediated phenomenon. A retrospective case-control study was undertaken to test the hypothesis that EER is more common among children with typical vasovagal syncope (VVS) than among their peers with nonvagal syncope (NVS) or with no syncope. Methods: Patients aged 4-18 years with syncope were identified by a single-centre database search followed by a review of history for features of VVS (n = 150) or NVS (n = 84). The first available electrocardiogram (ECG) for VVS or for NVS was retrieved. Age- and sex-matched children with no known syncope or heart disease were then identified (n = 216). ECGs were assessed separately for EER based on published criteria by 2 observers blinded to patients' clinical status. Results: Mean age was 12.3 ± 3.2 years, and heart rate was 74.2 ± 16.5 beats/min. EER was more prevalent in VVS (33.3%) than among patients with NVS (19.1%; odds ratio: 2.29; confidence interval: 1.32-5.50) or among those with no syncope (12.5%; odds ratio: 3.14; confidence interval: 1.81-5.46). Heart rates were significantly lower in VVS and NVS (heart rate: 70.1 ± 13.8 and 70.7 ± 12.4 beats/min, respectively) compared with children with no syncope (heart rate: 78.2 ± 18.0 beats/min), both P < 0.001. Conclusions: EER is more common in paediatric patients with VVS than those with NVS or without syncope, consistent with a possible vagal contribution to the ECG finding.
Contexte: La repolarisation précoce (RP) à l'électrocardiogramme (ECG) est liée à une fibrillation ventriculaire idiopathique chez les adultes. Fréquente chez les enfants, sa signification est toutefois nébuleuse. Certaines données laissent penser qu'il pourrait s'agir d'un phénomène d'origine vagale. Une étude rétrospective cas-témoins a donc été réalisée dans le but de vérifier l'hypothèse selon laquelle la RP à l'ECG est plus courante chez les enfants atteints de syncope vasovagale (SVV) typique que chez leurs pairs atteints de syncope non vagale (SNV) ou non atteints de syncope. Méthodologie: Des patients de 4 à 18 ans atteints de syncope ont été recensés au moyen d'une recherche dans la base de données d'un centre, suivie d'un examen des antécédents visant à retracer des manifestations de SVV (n = 150) ou de SNV (n = 84). Le premier ECG disponible traduisant une SVV ou une SNV a été récupéré. Un appariement selon l'âge et le sexe entre les sujets atteints et des enfants qui n'étaient pas atteints de syncope ni de maladie cardiaque (n = 216) a ensuite été effectué. Deux observateurs qui ne connaissaient pas l'état clinique des enfants ont évalué les ECG séparément, à la recherche d'une RP, en se basant sur les critères publiés. Résultats: L'âge moyen des sujets était de 12,3 ± 3,2 ans et la fréquence cardiaque moyenne, de 74,2 ± 16,5 battements/minute. La prévalence de la RP à l'ECG était plus élevée chez les patients atteints de SVV (33,3 %) que chez les patients atteints de SNV (19,1 %; rapport de cotes [RC] : 2,29; intervalle de confiance [IC] : 1,32-5,50) ou les enfants non atteints de syncope (12,5 %; RC : 3,14; IC : 1,81-5,46). La fréquence cardiaque (FC) était significativement plus faible chez les sujets atteints de SVV ou de SNV (FC : 70,1 ± 13,8 et 70,7 ± 12,4 battements/minute, respectivement), en comparaison des enfants non atteints de syncope (FC : 78,2 ± 18,0 battements/minute); p < 0,001 dans les deux cas. Conclusion: La repolarisation précoce à l'ECG est plus courante chez les enfants atteints de syncope vasovagale que chez les enfants atteints de syncope non vagale ou non atteints de syncope, ce qui concorde avec une possible composante vagale dans le tracé de l'ECG.
RESUMEN
BACKGROUND: Prognosticating outcomes for traumatic brain injury (TBI) patients is challenging due to the required specialized skills and variability among clinicians. Recent attempts to standardize TBI prognosis have leveraged machine learning (ML) methodologies. This study evaluates the necessity and influence of ML-assisted TBI prognostication through healthcare professionals' perspectives via focus group discussions. METHODS: Two virtual focus groups included ten key TBI care stakeholders (one neurosurgeon, two emergency clinicians, one internist, two radiologists, one registered nurse, two researchers in ML and healthcare and one patient representative). They answered six open-ended questions about their perceptions and potential ML use in TBI prognostication. Transcribed focus group discussions were thematically analyzed using qualitative data analysis software. RESULTS: The study captured diverse perceptions and interests in TBI prognostication across clinical specialties. Notably, certain clinicians who currently do not prognosticate expressed an interest in doing so independently provided they had access to ML support. Concerns included ML's accuracy and the need for proficient ML researchers in clinical settings. The consensus suggested using ML as a secondary consultation tool and promoting collaboration with internal or external research resources. Participants believed ML prognostication could enhance disposition planning and standardize care regardless of clinician expertise or injury severity. There was no evidence of perceived bias or interference during the discussions. CONCLUSION: Our findings revealed an overall positive attitude toward ML-based prognostication. Despite raising multiple concerns, the focus group discussions were particularly valuable in underscoring the potential of ML in democratizing and standardizing TBI prognosis practices.
RESUMEN
Computed tomography (CT) is an important imaging modality for guiding prognostication in patients with traumatic brain injury (TBI). However, because of the specialized expertise necessary, timely and dependable TBI prognostication based on CT imaging remains challenging. This study aimed to enhance the efficiency and reliability of TBI prognostication by employing machine learning (ML) techniques on CT images. A retrospective analysis was conducted on the Collaborative European NeuroTrauma Effectiveness Research in TBI (CENTER-TBI) data set (n = 1016). An ML-driven binary classifier was developed to predict favorable or unfavorable outcomes at 6 months post-injury. The prognostic performance was assessed using the area under the curve (AUC) over fivefold cross-validation and compared with conventional models that depend on clinical variables and CT scoring systems. An external validation was performed using the Comparative Indian Neurotrauma Effectiveness Research in Traumatic Brain Injury (CINTER-TBI) data set (n = 348). The developed model achieved superior performance without the necessity for manual CT assessments (AUC = 0.846 [95% CI: 0.843-0.849]) compared with the model based on the clinical and laboratory variables (AUC = 0.817 [95% CI: 0.814-0.820]) and established CT scoring systems requiring manual interpretations (AUC = 0.829 [95% CI: 0.826-0.832] for Marshall and 0.838 [95% CI: 0.835-0.841] for International Mission for Prognosis and Analysis of Clinical Trials in TBI [IMPACT]). The external validation demonstrated the prognostic capacity of the developed model to be significantly better (AUC = 0.859 [95% CI: 0.857-0.862]) than the model using clinical variables (AUC = 0.809 [95% CI: 0.798-0.820]). This study established an ML-based model that provides efficient and reliable TBI prognosis based on CT scans, with potential implications for earlier intervention and improved patient outcomes.
Asunto(s)
Lesiones Traumáticas del Encéfalo , Aprendizaje Automático , Tomografía Computarizada por Rayos X , Humanos , Lesiones Traumáticas del Encéfalo/diagnóstico por imagen , Masculino , Femenino , Pronóstico , Adulto , Persona de Mediana Edad , Tomografía Computarizada por Rayos X/métodos , Tomografía Computarizada por Rayos X/normas , Estudios Retrospectivos , Adulto Joven , Anciano , AdolescenteRESUMEN
Neuroimaging has a key role in identifying small-vessel vasculitis from common diseases it mimics, such as multiple sclerosis. Oftentimes, a multitude of these conditions present similarly, and thus diagnosis is difficult. To date, there is no standardized method to differentiate between these diseases. This review identifies and presents existing scoring tools that could serve as a starting point for integrating artificial intelligence/machine learning (AI/ML) into the clinical decision-making process for these rare diseases. A scoping literature review of EMBASE and MEDLINE included 114 articles to evaluate what criteria exist to diagnose small-vessel vasculitis and common mimics. This paper presents the existing criteria of small-vessel vasculitis conditions and mimics them to guide the future integration of AI/ML algorithms to aid in diagnosing these conditions, which present similarly and non-specifically.
Asunto(s)
Inteligencia Artificial , Vasculitis , Humanos , Aprendizaje Automático , Vasculitis/diagnóstico por imagen , Neuroimagen , Sistema Nervioso CentralRESUMEN
This scoping review examines the emerging field of synthetic ultrasound generation using machine learning (ML) models in radiology. Nineteen studies were analyzed, revealing three primary methodological strategies: unconditional generation, conditional generation, and domain translation. Synthetic ultrasound is mainly used to augment training datasets and as training material for radiologists. Blind expert assessment and Fréchet Inception Distance are common evaluation methods. Current limitations include the need for large training datasets, manual annotations for controllable generation, and insufficient research on incorporating new domain knowledge. While generative ultrasound models show promise, further work is required for clinical implementation.
Asunto(s)
Aprendizaje Automático , Radiología , Humanos , Ultrasonografía , Radiólogos , Procesamiento de Imagen Asistido por ComputadorRESUMEN
OBJECTIVES: This scoping review was conducted to determine the barriers and enablers associated with the acceptance of artificial intelligence/machine learning (AI/ML)-enabled innovations into radiology practice from a physician's perspective. METHODS: A systematic search was performed using Ovid Medline and Embase. Keywords were used to generate refined queries with the inclusion of computer-aided diagnosis, artificial intelligence, and barriers and enablers. Three reviewers assessed the articles, with a fourth reviewer used for disagreements. The risk of bias was mitigated by including both quantitative and qualitative studies. RESULTS: An electronic search from January 2000 to 2023 identified 513 studies. Twelve articles were found to fulfill the inclusion criteria: qualitative studies (n = 4), survey studies (n = 7), and randomized controlled trials (RCT) (n = 1). Among the most common barriers to AI implementation into radiology practice were radiologists' lack of acceptance and trust in AI innovations; a lack of awareness, knowledge, and familiarity with the technology; and perceived threat to the professional autonomy of radiologists. The most important identified AI implementation enablers were high expectations of AI's potential added value; the potential to decrease errors in diagnosis; the potential to increase efficiency when reaching a diagnosis; and the potential to improve the quality of patient care. CONCLUSIONS: This scoping review found that few studies have been designed specifically to identify barriers and enablers to the acceptance of AI in radiology practice. The majority of studies have assessed the perception of AI replacing radiologists, rather than other barriers or enablers in the adoption of AI. To comprehensively evaluate the potential advantages and disadvantages of integrating AI innovations into radiology practice, gathering more robust research evidence on stakeholder perspectives and attitudes is essential.
Asunto(s)
Radiología , Humanos , Inteligencia Artificial , Aprendizaje AutomáticoRESUMEN
BACKGROUND: Current artificial intelligence studies for supporting CT screening tasks depend on either supervised learning or detecting anomalies. However, the former involves a heavy annotation workload owing to requiring many slice-wise annotations (ground truth labels); the latter is promising, but while it reduces the annotation workload, it often suffers from lower performance. This study presents a novel weakly supervised anomaly detection (WSAD) algorithm trained based on scan-wise normal and anomalous annotations to provide better performance than conventional methods while reducing annotation workload. METHODS: Based on surveillance video anomaly detection methodology, feature vectors representing each CT slice were trained on an AR-Net-based convolutional network using a dynamic multiple-instance learning loss and a center loss function. The following two publicly available CT datasets were retrospectively analyzed: the RSNA brain hemorrhage dataset (normal scans: 12,862; scans with intracranial hematoma: 8882) and COVID-CT set (normal scans: 282; scans with COVID-19: 95). RESULTS: Anomaly scores of each slice were successfully predicted despite inaccessibility to any slice-wise annotations. Slice-level area under the curve (AUC), sensitivity, specificity, and accuracy from the brain CT dataset were 0.89, 0.85, 0.78, and 0.79, respectively. The proposed method reduced the number of annotations in the brain dataset by 97.1% compared to an ordinary slice-level supervised learning method. CONCLUSION: This study demonstrated a significant annotation reduction in identifying anomalous CT slices compared to a supervised learning approach. The effectiveness of the proposed WSAD algorithm was verified through higher AUC than existing anomaly detection techniques.
RESUMEN
Background: Training machine learning (ML) models in medical imaging requires large amounts of labeled data. To minimize labeling workload, it is common to divide training data among multiple readers for separate annotation without consensus and then combine the labeled data for training a ML model. This can lead to a biased training dataset and poor ML algorithm prediction performance. The purpose of this study is to determine if ML algorithms can overcome biases caused by multiple readers' labeling without consensus. Methods: This study used a publicly available chest X-ray dataset of pediatric pneumonia. As an analogy to a practical dataset without labeling consensus among multiple readers, random and systematic errors were artificially added to the dataset to generate biased data for a binary-class classification task. The Resnet18-based convolutional neural network (CNN) was used as a baseline model. A Resnet18 model with a regularization term added as a loss function was utilized to examine for improvement in the baseline model. Results: The effects of false positive labels, false negative labels, and random errors (5-25%) resulted in a loss of AUC (0-14%) when training a binary CNN classifier. The model with a regularized loss function improved the AUC (75-84%) over that of the baseline model (65-79%). Conclusion: This study indicated that it is possible for ML algorithms to overcome individual readers' biases when consensus is not available. It is recommended to use regularized loss functions when allocating annotation tasks to multiple readers as they are easy to implement and effective in mitigating biased labels.
Asunto(s)
Aprendizaje Automático , Redes Neurales de la Computación , Humanos , Niño , Algoritmos , SesgoRESUMEN
Supervised machine learning classification is the most common example of artificial intelligence (AI) in industry and in academic research. These technologies predict whether a series of measurements belong to one of multiple groups of examples on which the machine was previously trained. Prior to real-world deployment, all implementations need to be carefully evaluated with hold-out validation, where the algorithm is tested on different samples than it was provided for training, in order to ensure the generalizability and reliability of AI models. However, established methods for performing hold-out validation do not assess the consistency of the mistakes that the AI model makes during hold-out validation. Here, we show that in addition to standard methods, an enhanced technique for performing hold-out validation-that also assesses the consistency of the sample-wise mistakes made by the learning algorithm-can assist in the evaluation and design of reliable and predictable AI models. The technique can be applied to the validation of any supervised learning classification application, and we demonstrate the use of the technique on a variety of example biomedical diagnostic applications, which help illustrate the importance of producing reliable AI models. The validation software created is made publicly available, assisting anyone developing AI models for any supervised classification application in the creation of more reliable and predictable technologies.
RESUMEN
BACKGROUND: Although MRI is a radiation-free imaging modality, it has historically been limited in lung imaging due to inherent technical restrictions. The aim of this study is to explore the performance of lung MRI in detecting solid and subsolid pulmonary nodules using T1 gradient-echo (GRE) (VIBE, Volumetric interpolated breath-hold examination), ultrashort time echo (UTE) and T2 Fast Spin Echo (HASTE, Half fourier Single-shot Turbo spin-Echo). METHODS: Patients underwent a lung MRI in a 3 T scanner as part of a prospective research project. A baseline Chest CT was obtained as part of their standard of care. Nodules were identified and measured on the baseline CT and categorized according to their density (solid and subsolid) and size (> 4 mm/ ≤ 4 mm). Nodules seen on the baseline CT were classified as present or absent on the different MRI sequences by two thoracic radiologists independently. Interobserver agreement was determined using the simple Kappa coefficient. Paired differences were compared using nonparametric Mann-Whitney U tests. The McNemar test was used to evaluate paired differences in nodule detection between MRI sequences. RESULTS: Thirty-six patients were prospectively enrolled. One hundred forty-nine nodules (100 solid/49 subsolid) with mean size 10.8 mm (SD = 9.4) were included in the analysis. There was substantial interobserver agreement (k = 0.7, p = 0.05). Detection for all nodules, solid and subsolid nodules was respectively; UTE: 71.8%/71.0%/73.5%; VIBE: 61.6%/65%/55.1%; HASTE 72.4%/72.2%/72.7%. Detection rate was higher for nodules > 4 mm in all groups: UTE 90.2%/93.4%/85.4%, VIBE 78.4%/88.5%/63.4%, HASTE 89.4%/93.8%/83.8%. Detection of lesions ≤4 mm was low for all sequences. UTE and HASTE performed significantly better than VIBE for detection of all nodules and subsolid nodules (diff = 18.4 and 17.6%, p = < 0.01 and p = 0.03, respectively). There was no significant difference between UTE and HASTE. There were no significant differences amongst MRI sequences for solid nodules. CONCLUSIONS: Lung MRI shows adequate performance for the detection of solid and subsolid pulmonary nodules larger than 4 mm and can serve as a promising radiation-free alternative to CT.
Asunto(s)
Neoplasias Pulmonares , Pulmón , Humanos , Estudios Prospectivos , Pulmón/diagnóstico por imagen , Pulmón/patología , Imagen por Resonancia Magnética/métodos , Tomografía Computarizada por Rayos X/métodos , Neoplasias Pulmonares/diagnóstico por imagen , Neoplasias Pulmonares/patologíaRESUMEN
BACKGROUND: The purpose of this study was to conduct a systematic review for understanding the availability and limitations of artificial intelligence (AI) approaches that could automatically identify and quantify computed tomography (CT) findings in traumatic brain injury (TBI). METHODS: Systematic review, in accordance with PRISMA 2020 and SPIRIT-AI extension guidelines, with a search of 4 databases (Medline, Embase, IEEE Xplore, and Web of Science) was performed to find AI studies that automated the clinical tasks for identifying and quantifying CT findings of TBI-related abnormalities. RESULTS: A total of 531 unique publications were reviewed, which resulted in 66 articles that met our inclusion criteria. The following components for identification and quantification regarding TBI were covered and automated by existing AI studies: identification of TBI-related abnormalities; classification of intracranial hemorrhage types; slice-, pixel-, and voxel-level localization of hemorrhage; measurement of midline shift; and measurement of hematoma volume. Automated identification of obliterated basal cisterns was not investigated in the existing AI studies. Most of the AI algorithms were based on deep neural networks that were trained on 2- or 3-dimensional CT imaging datasets. CONCLUSION: We identified several important TBI-related CT findings that can be automatically identified and quantified with AI. A combination of these techniques may provide useful tools to enhance reproducibility of TBI identification and quantification by supporting radiologists and clinicians in their TBI assessments and reducing subjective human factors.
Asunto(s)
Inteligencia Artificial , Lesiones Traumáticas del Encéfalo , Humanos , Reproducibilidad de los Resultados , Cintigrafía , Lesiones Traumáticas del Encéfalo/diagnóstico por imagen , Tomografía Computarizada por Rayos XRESUMEN
OBJECTIVES: Unlike in adult rheumatology, for most forms of juvenile idiopathic arthritis (JIA) no reliable biomarkers currently exist to assess joint and disease activity. However, electrophoresis is frequently found changed in active juvenile arthritis. The objective of this study was to evaluate the α2-fraction of serum electrophoresis and its main components as biomarkers for JIA, categories extended/persistent oligoarthritis and seronegative polyarthritis, in comparison with the conventionally used erythrocyte sedimentation rate and C-reactive protein. METHODS: Serum samples and clinical data from 181 patients with JIA were collected. Serum electrophoresis and α2-fraction and its components were determined using standard methods. Relationship between calculated α2-fraction of serum electrophoresis (CA2F) and its components, acute-phase parameters and cJADAS27 was assessed using Pearson's correlation coefficient and linear regression modelling, adjusting for confounding effects. Results were confirmed in a second cohort with 223 serum samples from 37 patients, using a mixed model to account for repeated measures. RESULTS: Compared to ESR and CRP, CA2F showed higher correlation to cJADAS27, in particular for persistent oligoarthritis. Of the three components of the α2-fraction, haptoglobin showed the highest correlation to cJADAS27. Regression analysis demonstrated higher ability to predict cJADAS27 for CA2F, and especially for haptoglobin as a component thereof, than for CRP and ESR. CONCLUSION: Compared to conventional methods, α2-fraction of serum electrophoresis and specifically, haptoglobin show higher correlations with disease activity in common subtypes of JIA, representing excellent candidates as biomarkers for disease activity. Further studies are necessary to determine diagnostic value and correlations in other subtypes.
Asunto(s)
Artritis Juvenil , Biomarcadores , Sedimentación Sanguínea , Proteína C-Reactiva/análisis , Haptoglobinas/análisis , HumanosRESUMEN
Background: Organ stiffening can be caused by inflammation and fibrosis, processes that are common causes of transplant kidney dysfunction. Magnetic resonance elastography (MRE) is a contrast-free, noninvasive imaging modality that measures kidney stiffness. The objective of this study was to assess the ability of MRE to serve as a prognostic factor for renal outcomes. Methods: Patients were recruited from the St Michael's Hospital Kidney Transplant Clinic. Relevant baseline demographic, clinical, and Banff histologic information, along with follow-up estimated glomerular filtration rate (eGFR) data, were recorded. Two-dimensional gradient-echo MRE imaging was performed to obtain kidney "stiffness" maps. Binary logistic regression analyses were performed to examine for relationships between stiffness and microvascular inflammation score. Linear mixed-effects modeling was used to assess the relationship between stiffness and eGFR change over time controlling for other baseline variables. A G2-likelihood ratio Chi-squared test was performed to compare between the baseline models with and without "stiffness." Results: Sixty-eight transplant kidneys were scanned in 66 patients (mean age 56 ± 12 y, 24 females), with 38 allografts undergoing a contemporaneous biopsy. Mean transplant vintage was 7.0 ± 6.8 y. In biopsied allografts, MRE-derived allograft stiffness was associated only with microvascular inflammation (Banff g + ptc score, Spearman ρ = 0.43, P = 0.01), but no other histologic parameters. Stiffness was negatively associated with eGFR change over time (Stiffness × Time interaction ß = -0.80, P < 0.0001), a finding that remained significant even when adjusted for biopsy status and baseline variables (Stiffness × Time interaction ß = -0.46, P = 0.04). Conversely, the clinical models including "stiffness" showed significantly better fit (P = 0.04) compared with the baseline clinical models without "stiffness." Conclusions: MRE-derived renal stiffness provides important prognostic information regarding renal function loss for patients with allograft dysfunction, over and above what is provided by current clinical variables.
RESUMEN
Background The Ovarian-Adnexal Reporting and Data System (O-RADS) US risk stratification and management system (O-RADS US) was designed to improve risk assessment and management of ovarian and adnexal lesions. Validation studies including both surgical and nonsurgical treatment as the reference standard remain lacking. Purpose To externally validate O-RADS US in women who underwent either surgical or nonsurgical treatment and to determine if incorporating acoustic shadowing as a benign finding improves diagnostic performance. Materials and Methods This retrospective study included consecutive women who underwent pelvic US between August 2015 and April 2017 at a tertiary referral oncology center. Two independent readers blinded to clinical and histologic outcome assigned an O-RADS risk category and an International Ovarian Tumor Analysis (IOTA) Assessment of Different NEoplasias in the adneXa (ADNEX) model risk of malignancy score to assessable lesions. Reference standards were surgical histopathology or 2-year imaging follow-up. Receiver operating characteristic (ROC) curve analysis was used to evaluate performance of the O-RADS US, ADNEX, and modified O-RADS models incorporating acoustic shadowing. Results In total, 227 women (mean age, 52 years ± 16 [SD]) with 262 ovarian or adnexal lesions were evaluated. Of these lesions, 187 (71%) were benign and 75 (29%) were malignant. The proportion of malignancy was 0% (0 of 100) for O-RADS 2, 3% (one of 32) for O-RADS 3, 35% (22 of 63) for O-RADS 4, and 78% (52 of 67) for O-RADS 5. The area under the ROC curve (AUC) for O-RADS and ADNEX was 0.91 (95% CI: 0.88, 0.94) and 0.95 (95% CI: 0.92, 0.97; P = .01), respectively. The addition of acoustic shadowing as a benign finding improved O-RADS AUC to 0.94 (95% CI: 0.91, 0.96; P = .01). Use of O-RADS 4 as a threshold yielded a sensitivity of 99% (74 of 75; 95% CI: 96, 100) and a specificity of 70% (131 of 187; 95% CI: 64, 77). Conclusion In a tertiary referral oncology center, the Ovarian-Adnexal Reporting and Data System US risk stratification and management system enabled accurate distinction of benign from malignant ovarian and adnexal lesions. Adding acoustic shadowing as a benign finding improved its diagnostic performance. © RSNA, 2022 See also the editorial by Levine in this issue.
Asunto(s)
Enfermedades de los Anexos , Neoplasias Ováricas , Enfermedades de los Anexos/patología , Sistemas de Datos , Femenino , Humanos , Persona de Mediana Edad , Neoplasias Ováricas/patología , Estudios Retrospectivos , Medición de Riesgo , Sensibilidad y Especificidad , Ultrasonografía/métodosRESUMEN
PURPOSE: Machine learning (ML) models in medical imaging (MI) can be of great value in computer aided diagnostic systems, but little attention is given to the confidence (alternatively, uncertainty) of such models, which may have significant clinical implications. This paper applied, validated, and explored a technique for assessing uncertainty in convolutional neural networks (CNNs) in the context of MI. MATERIALS AND METHODS: We used two publicly accessible imaging datasets: a chest x-ray dataset (pneumonia vs. control) and a skin cancer imaging dataset (malignant vs. benign) to explore the proposed measure of uncertainty based on experiments with different class imbalance-sample sizes, and experiments with images close to the classification boundary. We also further verified our hypothesis by examining the relationship with other performance metrics and cross-checking CNN predictions and confidence scores with an expert radiologist (available in the Supplementary Information). Additionally, bounds were derived on the uncertainty metric, and recommendations for interpretability were made. RESULTS: With respect to training set class imbalance for the pneumonia MI dataset, the uncertainty metric was minimized when both classes were nearly equal in size (regardless of training set size) and was approximately 17% smaller than the maximum uncertainty resulting from greater imbalance. We found that less-obvious test images (those closer to the classification boundary) produced higher classification uncertainty, about 10-15 times greater than images further from the boundary. Relevant MI performance metrics like accuracy, sensitivity, and sensibility showed seemingly negative linear correlations, though none were statistically significant (p [Formula: see text] 0.05). The expert radiologist and CNN expressed agreement on a small sample of test images, though this finding is only preliminary. CONCLUSIONS: This paper demonstrated the importance of uncertainty reporting alongside predictions in medical imaging. Results demonstrate considerable potential from automatically assessing classifier reliability on each prediction with the proposed uncertainty metric.
Asunto(s)
Aprendizaje Automático , Redes Neurales de la Computación , Diagnóstico por Imagen , Humanos , Reproducibilidad de los Resultados , IncertidumbreRESUMEN
Purpose: To determine if MRI features and molecular subtype influence the detectability of breast cancers on MRI in high-risk patients. Methods and Materials: Breast cancers in a high-risk population of 104 patients were diagnosed following MRI describing a BI-RADS 4-5 lesion. MRI characteristics at the time of diagnosis were compared with previous MRI, where a BI-RADS 1-2-3 lesion was described. Results: There were 77 false-negative MRIs. A total of 51 cancers were overlooked and 26 were misinterpreted. There was no association found between MRI characteristics, the receptor type and the frequency of missed cancers. The main factors for misinterpreted lesions were multiple breast lesions, prior biopsy/surgery and long-term stability. Lesions were mostly overlooked because of their small size and high background parenchymal enhancement. Among missed lesions, 50% of those with plateau kinetics on initial MRI changed for washout kinetics, and 65% of initially progressively enhancing lesions then showed plateau or washout kinetics. There were more basal-like tumours in BRCA1 carriers (50%) than in non-carriers (13%), p = 0.0001, OR = 6.714, 95% CI = [2.058-21.910]. The proportion of missed cancers was lower in BRCA carriers (59%) versus non-carriers (79%), p < 0.05, OR = 2.621, 95% CI = [1.02-6.74]. Conclusions: MRI characteristics or molecular subtype do not influence breast cancer detectability. Lesions in a post-surgical breast should be assessed with caution. Long-term stability does not rule out malignancy and multimodality evaluation is of added value. Lowering the biopsy threshold for lesions with an interval change in kinetics for a type 2 or 3 curve should be considered. There was a higher rate of interval cancers in BRCA 1 patients attributed to lesions more aggressive in nature.
Asunto(s)
Neoplasias de la Mama , Mama/diagnóstico por imagen , Neoplasias de la Mama/diagnóstico por imagen , Neoplasias de la Mama/patología , Estudios de Casos y Controles , Femenino , Humanos , Imagen por Resonancia Magnética/métodos , Estudios RetrospectivosRESUMEN
ABSTRACT: Guest, NS, Corey, P, Tyrrell, PN, and El-Sohemy, A. Effect of caffeine on endurance performance in athletes may depend on HTR2A and CYP1A2 genotypes. J Strength Cond Res 36(9): 2486-2492, 2022-This investigation determined whether variation in the HTR2A (serotonin receptor) gene modifies the ergogenic effects of caffeine on endurance and further modifies performance by the CYP1A2 genotype. Male athletes ( n = 100; 25 ± 4 years) completed 10-km cycling time trials under 3 conditions as follows: 0, 2, or 4 mg of caffeine per kg body mass. Using a randomized, double-blinded, placebo-controlled design, data were analyzed using analysis of covariance to compare changes in cycling time between placebo (0 mg·kg -1 ) and each caffeine dose and adjusted for the placebo trial and order of treatment. A significance of ρ ≤ 0.05 was used. Subjects were genotyped for HTR2A (rs6313) and CYP1A2 (rs762551). A significant caffeine- HTR2A interaction ( p = 0.003) was observed; however, after adjustment for placebo trials, the interaction was no longer significant ( p = 0.37). Because of the strong caffeine- CYP1A2 interaction ( p < 0.0001) previously reported in these subjects, where the 4-mg dose resulted in divergent effects (slower and faster) on the 10-km cycling time, we conducted a simplified model to examine these same factors by the HTR2A genotype. The post hoc analysis excluded HTR2A CT heterozygotes and 2-mg·kg -1 caffeine trials. Among CYP1A2 fast metabolizers alone, a significant difference (1.7 minutes; p = 0.006) was observed when comparing (4- vs. 0-mg·kg -1 caffeine trials) between the HTR2A CC ( n = 16; 2.4 minutes) and TT ( n = 7; 0.7 minutes) genotypes. Our results show that 4-mg·kg -1 caffeine improves performance in individuals with the HTR2A CC genotype but only in those who are also CYP1A2 AA fast metabolizers. This study was registered with clinicaltrials.gov (NCT02109783).
Asunto(s)
Atletas , Cafeína , Citocromo P-450 CYP1A2 , Sustancias para Mejorar el Rendimiento , Receptor de Serotonina 5-HT2A , Cafeína/farmacología , Citocromo P-450 CYP1A2/genética , Método Doble Ciego , Genotipo , Humanos , Masculino , Sustancias para Mejorar el Rendimiento/farmacología , Receptor de Serotonina 5-HT2A/genéticaRESUMEN
PURPOSE: The effect of caffeine on anaerobic performance is unclear and may differ depending on an individual's genetics. The goal of this study was to determine whether caffeine influences anaerobic performance in a 30 s Wingate test, and if 14 single nucleotide polymorphisms (SNPs) in nine genes, associated with caffeine metabolism or response, modify caffeine's effects. METHODS: Competitive male athletes (N = 100; 25 ± 4 years) completed the Wingate under three conditions: 0, 2, or 4 mg of caffeine per kg of body mass (mg kg-1), using a double-blinded, placebo-controlled design. Using saliva samples, participants were genotyped for the 14 SNPs. The outcomes were peak power (Watts [W]), average power (Watts [W]), and fatigue index (%). RESULTS: There was no main effect of caffeine on Wingate outcomes. One significant caffeine-gene interaction was observed for CYP1A2 (rs762551, p = 0.004) on average power. However, post hoc analysis showed no difference in caffeine's effects within CYP1A2 genotypes for average power performance. No significant caffeine-gene interactions were observed for the remaining SNPs on peak power, average power and fatigue index. CONCLUSION: Caffeine had no effect on anaerobic performance and variations in several genes did not modify any effects of caffeine. TRIAL REGISTRATION: This study was registered with clinicaltrials.gov (NCT02109783).