RESUMEN
Purpose To develop, externally test, and evaluate clinical acceptability of a deep learning pediatric brain tumor segmentation model using stepwise transfer learning. Materials and Methods In this retrospective study, the authors leveraged two T2-weighted MRI datasets (May 2001 through December 2015) from a national brain tumor consortium (n = 184; median age, 7 years [range, 1-23 years]; 94 male patients) and a pediatric cancer center (n = 100; median age, 8 years [range, 1-19 years]; 47 male patients) to develop and evaluate deep learning neural networks for pediatric low-grade glioma segmentation using a stepwise transfer learning approach to maximize performance in a limited data scenario. The best model was externally tested on an independent test set and subjected to randomized blinded evaluation by three clinicians, wherein they assessed clinical acceptability of expert- and artificial intelligence (AI)-generated segmentations via 10-point Likert scales and Turing tests. Results The best AI model used in-domain stepwise transfer learning (median Dice score coefficient, 0.88 [IQR, 0.72-0.91] vs 0.812 [IQR, 0.56-0.89] for baseline model; P = .049). With external testing, the AI model yielded excellent accuracy using reference standards from three clinical experts (median Dice similarity coefficients: expert 1, 0.83 [IQR, 0.75-0.90]; expert 2, 0.81 [IQR, 0.70-0.89]; expert 3, 0.81 [IQR, 0.68-0.88]; mean accuracy, 0.82). For clinical benchmarking (n = 100 scans), experts rated AI-based segmentations higher on average compared with other experts (median Likert score, 9 [IQR, 7-9] vs 7 [IQR 7-9]) and rated more AI segmentations as clinically acceptable (80.2% vs 65.4%). Experts correctly predicted the origin of AI segmentations in an average of 26.0% of cases. Conclusion Stepwise transfer learning enabled expert-level automated pediatric brain tumor autosegmentation and volumetric measurement with a high level of clinical acceptability. Keywords: Stepwise Transfer Learning, Pediatric Brain Tumors, MRI Segmentation, Deep Learning Supplemental material is available for this article. © RSNA, 2024.
Asunto(s)
Neoplasias Encefálicas , Aprendizaje Profundo , Imagen por Resonancia Magnética , Humanos , Niño , Neoplasias Encefálicas/diagnóstico por imagen , Neoplasias Encefálicas/patología , Imagen por Resonancia Magnética/métodos , Masculino , Adolescente , Preescolar , Estudios Retrospectivos , Femenino , Lactante , Adulto Joven , Glioma/diagnóstico por imagen , Glioma/patología , Interpretación de Imagen Asistida por Computador/métodosRESUMEN
Pediatric glioma recurrence can cause morbidity and mortality; however, recurrence pattern and severity are heterogeneous and challenging to predict with established clinical and genomic markers. Resultingly, almost all children undergo frequent, long-term, magnetic resonance (MR) brain surveillance regardless of individual recurrence risk. Deep learning analysis of longitudinal MR may be an effective approach for improving individualized recurrence prediction in gliomas and other cancers but has thus far been infeasible with current frameworks. Here, we propose a self-supervised, deep learning approach to longitudinal medical imaging analysis, temporal learning, that models the spatiotemporal information from a patient's current and prior brain MRs to predict future recurrence. We apply temporal learning to pediatric glioma surveillance imaging for 715 patients (3,994 scans) from four distinct clinical settings. We find that longitudinal imaging analysis with temporal learning improves recurrence prediction performance by up to 41% compared to traditional approaches, with improvements in performance in both low- and high-grade glioma. We find that recurrence prediction accuracy increases incrementally with the number of historical scans available per patient. Temporal deep learning may enable point-of-care decision-support for pediatric brain tumors and be adaptable more broadly to patients with other cancers and chronic diseases undergoing surveillance imaging.
RESUMEN
Unenhanced CT scans exhibit high specificity in detecting moderate-to-severe hepatic steatosis. Even though many CTs are scanned from health screening and various diagnostic contexts, their potential for hepatic steatosis detection has largely remained unexplored. The accuracy of previous methodologies has been limited by the inclusion of non-parenchymal liver regions. To overcome this limitation, we present a novel deep-learning (DL) based method tailored for the automatic selection of parenchymal portions in CT images. This innovative method automatically delineates circular regions for effectively detecting hepatic steatosis. We use 1,014 multinational CT images to develop a DL model for segmenting liver and selecting the parenchymal regions. The results demonstrate outstanding performance in both tasks. By excluding non-parenchymal portions, our DL-based method surpasses previous limitations, achieving radiologist-level accuracy in liver attenuation measurements and hepatic steatosis detection. To ensure the reproducibility, we have openly shared 1014 annotated CT images and the DL system codes. Our novel research contributes to the refinement the automated detection methodologies of hepatic steatosis on CT images, enhancing the accuracy and efficiency of healthcare screening processes.
Asunto(s)
Aprendizaje Profundo , Hígado Graso , Hígado , Tomografía Computarizada por Rayos X , Humanos , Tomografía Computarizada por Rayos X/métodos , Hígado Graso/diagnóstico por imagen , Hígado Graso/patología , Hígado/diagnóstico por imagen , Hígado/patología , Masculino , Reproducibilidad de los Resultados , FemeninoRESUMEN
Purpose To develop and externally test a scan-to-prediction deep learning pipeline for noninvasive, MRI-based BRAF mutational status classification for pediatric low-grade glioma. Materials and Methods This retrospective study included two pediatric low-grade glioma datasets with linked genomic and diagnostic T2-weighted MRI data of patients: Dana-Farber/Boston Children's Hospital (development dataset, n = 214 [113 (52.8%) male; 104 (48.6%) BRAF wild type, 60 (28.0%) BRAF fusion, and 50 (23.4%) BRAF V600E]) and the Children's Brain Tumor Network (external testing, n = 112 [55 (49.1%) male; 35 (31.2%) BRAF wild type, 60 (53.6%) BRAF fusion, and 17 (15.2%) BRAF V600E]). A deep learning pipeline was developed to classify BRAF mutational status (BRAF wild type vs BRAF fusion vs BRAF V600E) via a two-stage process: (a) three-dimensional tumor segmentation and extraction of axial tumor images and (b) section-wise, deep learning-based classification of mutational status. Knowledge-transfer and self-supervised approaches were investigated to prevent model overfitting, with a primary end point of the area under the receiver operating characteristic curve (AUC). To enhance model interpretability, a novel metric, center of mass distance, was developed to quantify the model attention around the tumor. Results A combination of transfer learning from a pretrained medical imaging-specific network and self-supervised label cross-training (TransferX) coupled with consensus logic yielded the highest classification performance with an AUC of 0.82 (95% CI: 0.72, 0.91), 0.87 (95% CI: 0.61, 0.97), and 0.85 (95% CI: 0.66, 0.95) for BRAF wild type, BRAF fusion, and BRAF V600E, respectively, on internal testing. On external testing, the pipeline yielded an AUC of 0.72 (95% CI: 0.64, 0.86), 0.78 (95% CI: 0.61, 0.89), and 0.72 (95% CI: 0.64, 0.88) for BRAF wild type, BRAF fusion, and BRAF V600E, respectively. Conclusion Transfer learning and self-supervised cross-training improved classification performance and generalizability for noninvasive pediatric low-grade glioma mutational status prediction in a limited data scenario. Keywords: Pediatrics, MRI, CNS, Brain/Brain Stem, Oncology, Feature Detection, Diagnosis, Supervised Learning, Transfer Learning, Convolutional Neural Network (CNN) Supplemental material is available for this article. © RSNA, 2024.
Asunto(s)
Neoplasias Encefálicas , Glioma , Humanos , Niño , Masculino , Femenino , Neoplasias Encefálicas/diagnóstico por imagen , Estudios Retrospectivos , Proteínas Proto-Oncogénicas B-raf/genética , Glioma/diagnóstico , Aprendizaje AutomáticoRESUMEN
Lean muscle mass (LMM) is an important aspect of human health. Temporalis muscle thickness is a promising LMM marker but has had limited utility due to its unknown normal growth trajectory and reference ranges and lack of standardized measurement. Here, we develop an automated deep learning pipeline to accurately measure temporalis muscle thickness (iTMT) from routine brain magnetic resonance imaging (MRI). We apply iTMT to 23,876 MRIs of healthy subjects, ages 4 through 35, and generate sex-specific iTMT normal growth charts with percentiles. We find that iTMT was associated with specific physiologic traits, including caloric intake, physical activity, sex hormone levels, and presence of malignancy. We validate iTMT across multiple demographic groups and in children with brain tumors and demonstrate feasibility for individualized longitudinal monitoring. The iTMT pipeline provides unprecedented insights into temporalis muscle growth during human development and enables the use of LMM tracking to inform clinical decision-making.
Asunto(s)
Gráficos de Crecimiento , Músculo Temporal , Masculino , Femenino , Humanos , Niño , Músculo Temporal/diagnóstico por imagen , Músculo Temporal/patologíaRESUMEN
Importance: Sarcopenia is an established prognostic factor in patients with head and neck squamous cell carcinoma (HNSCC); the quantification of sarcopenia assessed by imaging is typically achieved through the skeletal muscle index (SMI), which can be derived from cervical skeletal muscle segmentation and cross-sectional area. However, manual muscle segmentation is labor intensive, prone to interobserver variability, and impractical for large-scale clinical use. Objective: To develop and externally validate a fully automated image-based deep learning platform for cervical vertebral muscle segmentation and SMI calculation and evaluate associations with survival and treatment toxicity outcomes. Design, Setting, and Participants: For this prognostic study, a model development data set was curated from publicly available and deidentified data from patients with HNSCC treated at MD Anderson Cancer Center between January 1, 2003, and December 31, 2013. A total of 899 patients undergoing primary radiation for HNSCC with abdominal computed tomography scans and complete clinical information were selected. An external validation data set was retrospectively collected from patients undergoing primary radiation therapy between January 1, 1996, and December 31, 2013, at Brigham and Women's Hospital. The data analysis was performed between May 1, 2022, and March 31, 2023. Exposure: C3 vertebral skeletal muscle segmentation during radiation therapy for HNSCC. Main Outcomes and Measures: Overall survival and treatment toxicity outcomes of HNSCC. Results: The total patient cohort comprised 899 patients with HNSCC (median [range] age, 58 [24-90] years; 140 female [15.6%] and 755 male [84.0%]). Dice similarity coefficients for the validation set (n = 96) and internal test set (n = 48) were 0.90 (95% CI, 0.90-0.91) and 0.90 (95% CI, 0.89-0.91), respectively, with a mean 96.2% acceptable rate between 2 reviewers on external clinical testing (n = 377). Estimated cross-sectional area and SMI values were associated with manually annotated values (Pearson r = 0.99; P < .001) across data sets. On multivariable Cox proportional hazards regression, SMI-derived sarcopenia was associated with worse overall survival (hazard ratio, 2.05; 95% CI, 1.04-4.04; P = .04) and longer feeding tube duration (median [range], 162 [6-1477] vs 134 [15-1255] days; hazard ratio, 0.66; 95% CI, 0.48-0.89; P = .006) than no sarcopenia. Conclusions and Relevance: This prognostic study's findings show external validation of a fully automated deep learning pipeline to accurately measure sarcopenia in HNSCC and an association with important disease outcomes. The pipeline could enable the integration of sarcopenia assessment into clinical decision making for individuals with HNSCC.
Asunto(s)
Aprendizaje Profundo , Neoplasias de Cabeza y Cuello , Sarcopenia , Humanos , Masculino , Femenino , Persona de Mediana Edad , Carcinoma de Células Escamosas de Cabeza y Cuello/diagnóstico por imagen , Estudios Retrospectivos , Sarcopenia/diagnóstico por imagen , Sarcopenia/complicaciones , Neoplasias de Cabeza y Cuello/complicaciones , Neoplasias de Cabeza y Cuello/diagnóstico por imagenRESUMEN
Purpose: To develop and externally validate a scan-to-prediction deep-learning pipeline for noninvasive, MRI-based BRAF mutational status classification for pLGG. Materials and Methods: We conducted a retrospective study of two pLGG datasets with linked genomic and diagnostic T2-weighted MRI of patients: BCH (development dataset, n=214 [60 (28%) BRAF fusion, 50 (23%) BRAF V600E, 104 (49%) wild-type), and Child Brain Tumor Network (CBTN) (external validation, n=112 [60 (53%) BRAF-Fusion, 17 (15%) BRAF-V600E, 35 (32%) wild-type]). We developed a deep learning pipeline to classify BRAF mutational status (V600E vs. fusion vs. wildtype) via a two-stage process: 1) 3D tumor segmentation and extraction of axial tumor images, and 2) slice-wise, deep learning-based classification of mutational status. We investigated knowledge-transfer and self-supervised approaches to prevent model overfitting with a primary endpoint of the area under the receiver operating characteristic curve (AUC). To enhance model interpretability, we developed a novel metric, COMDist, that quantifies the accuracy of model attention around the tumor. Results: A combination of transfer learning from a pretrained medical imaging-specific network and self-supervised label cross-training (TransferX) coupled with consensus logic yielded the highest macro-average AUC (0.82 [95% CI: 0.70-0.90]) and accuracy (77%) on internal validation, with an AUC improvement of +17.7% and a COMDist improvement of +6.4% versus training from scratch. On external validation, the TransferX model yielded AUC (0.73 [95% CI 0.68-0.88]) and accuracy (75%). Conclusion: Transfer learning and self-supervised cross-training improved classification performance and generalizability for noninvasive pLGG mutational status prediction in a limited data scenario.
RESUMEN
Artificial intelligence (AI) and machine learning (ML) are becoming critical in developing and deploying personalized medicine and targeted clinical trials. Recent advances in ML have enabled the integration of wider ranges of data including both medical records and imaging (radiomics). However, the development of prognostic models is complex as no modeling strategy is universally superior to others and validation of developed models requires large and diverse datasets to demonstrate that prognostic models developed (regardless of method) from one dataset are applicable to other datasets both internally and externally. Using a retrospective dataset of 2,552 patients from a single institution and a strict evaluation framework that included external validation on three external patient cohorts (873 patients), we crowdsourced the development of ML models to predict overall survival in head and neck cancer (HNC) using electronic medical records (EMR) and pretreatment radiological images. To assess the relative contributions of radiomics in predicting HNC prognosis, we compared 12 different models using imaging and/or EMR data. The model with the highest accuracy used multitask learning on clinical data and tumor volume, achieving high prognostic accuracy for 2-year and lifetime survival prediction, outperforming models relying on clinical data only, engineered radiomics, or complex deep neural network architecture. However, when we attempted to extend the best performing models from this large training dataset to other institutions, we observed significant reductions in the performance of the model in those datasets, highlighting the importance of detailed population-based reporting for AI/ML model utility and stronger validation frameworks. We have developed highly prognostic models for overall survival in HNC using EMRs and pretreatment radiological images based on a large, retrospective dataset of 2,552 patients from our institution.Diverse ML approaches were used by independent investigators. The model with the highest accuracy used multitask learning on clinical data and tumor volume.External validation of the top three performing models on three datasets (873 patients) with significant differences in the distributions of clinical and demographic variables demonstrated significant decreases in model performance. Significance: ML combined with simple prognostic factors outperformed multiple advanced CT radiomics and deep learning methods. ML models provided diverse solutions for prognosis of patients with HNC but their prognostic value is affected by differences in patient populations and require extensive validation.
Asunto(s)
Aprendizaje Profundo , Neoplasias de Cabeza y Cuello , Humanos , Pronóstico , Estudios Retrospectivos , Inteligencia Artificial , Neoplasias de Cabeza y Cuello/diagnóstico por imagenRESUMEN
Purpose: Artificial intelligence (AI)-automated tumor delineation for pediatric gliomas would enable real-time volumetric evaluation to support diagnosis, treatment response assessment, and clinical decision-making. Auto-segmentation algorithms for pediatric tumors are rare, due to limited data availability, and algorithms have yet to demonstrate clinical translation. Methods: We leveraged two datasets from a national brain tumor consortium (n=184) and a pediatric cancer center (n=100) to develop, externally validate, and clinically benchmark deep learning neural networks for pediatric low-grade glioma (pLGG) segmentation using a novel in-domain, stepwise transfer learning approach. The best model [via Dice similarity coefficient (DSC)] was externally validated and subject to randomized, blinded evaluation by three expert clinicians wherein clinicians assessed clinical acceptability of expert- and AI-generated segmentations via 10-point Likert scales and Turing tests. Results: The best AI model utilized in-domain, stepwise transfer learning (median DSC: 0.877 [IQR 0.715-0.914]) versus baseline model (median DSC 0.812 [IQR 0.559-0.888]; p<0.05). On external testing (n=60), the AI model yielded accuracy comparable to inter-expert agreement (median DSC: 0.834 [IQR 0.726-0.901] vs. 0.861 [IQR 0.795-0.905], p=0.13). On clinical benchmarking (n=100 scans, 300 segmentations from 3 experts), the experts rated the AI model higher on average compared to other experts (median Likert rating: 9 [IQR 7-9]) vs. 7 [IQR 7-9], p<0.05 for each). Additionally, the AI segmentations had significantly higher (p<0.05) overall acceptability compared to experts on average (80.2% vs. 65.4%). Experts correctly predicted the origins of AI segmentations in an average of 26.0% of cases. Conclusions: Stepwise transfer learning enabled expert-level, automated pediatric brain tumor auto-segmentation and volumetric measurement with a high level of clinical acceptability. This approach may enable development and translation of AI imaging segmentation algorithms in limited data scenarios.
RESUMEN
OBJECTIVES: Using diffusion basis spectrum imaging (DBSI) to examine the microstructural changes in the substantia nigra (SN) and global white matter (WM) tracts of patients with early-stage PD. METHODS: Thirty-seven age- and sex-matched patients with early-stage PD and 22 healthy controls (HCs) were enrolled in this study. All participants underwent clinical assessments and diffusion-weighted MRI scans, analyzed by diffusion tensor imaging (DTI) and DBSI to assess the pathologies of PD in SN and global WM tracts. RESULTS: The lower DTI fraction anisotropy (FA) was seen in SN of PD patients (PD: 0.316 ± 0.034 vs HCs: 0.331 ± 0.019, p = 0.015). The putative cells marker-DBSI-restricted fraction (PD: 0.132 ± 0.051 vs HCs: 0.105 ± 0.039, p = 0.031) and the edema/extracellular space marker-DBSI non-restricted-fraction (PD: 0.150 ± 0.052 vs HCs: 0.122 ± 0.052, p = 0.020) were both significantly higher and the density of axons/dendrites marker-DBSI fiber-fraction (PD: 0.718 ± 0.073 vs HCs: 0.773 ± 0.071, p = 0.003) was significantly lower in SN of PD patients. DBSI-restricted fraction in SN was negatively correlated with HAMA scores (r = - 0.501, p = 0.005), whereas DTI-FA was not correlated with any clinical scales. In WM tracts, only higher DTI axial diffusivity (AD) among DTI metrics was found in multiple WM regions in PD, while lower DBSI fiber-fraction and higher DBSI non-restricted-fraction were detected in multiple WM regions. DBSI non-restricted-fraction in both left fornix (cres)/stria terminalis (r = -0.472, p = 0.004) and right posterior thalamic radiation (r = - 0.467, p = 0.005) was negatively correlated with MMSE scores. CONCLUSION: DBSI could potentially detect and quantify the extent of inflammatory cell infiltration, fiber/dendrite loss, and edema in both SN and WM tracts in patients with early-stage PD, a finding remains to be further investigated through more extensive longitudinal DBSI analysis. CLINICAL RELEVANCE STATEMENT: Our study shows that DBSI indexes can potentially detect early-stage PD's pathological changes, with a notable ability to distinguish between inflammation and edema. This implies that DBSI has the potential to be an imaging biomarker for early PD diagnosis. KEY POINTS: ⢠Diffusion basis spectrum imaging detected higher restricted-fraction in Parkinson's disease, potentially reflecting inflammatory cell infiltration. ⢠Diffusion basis spectrum imaging detected higher non-restricted-fraction and lower fiber-fraction in Parkinson's disease, indicating the presence of edema and/or dopaminergic neuronal/dendritic loss. ⢠Diffusion basis spectrum imaging metrics correlated with non-motor symptoms, suggesting its potential diagnostic role to detect early-stage PD dysfunctions.
Asunto(s)
Enfermedad de Parkinson , Sustancia Blanca , Humanos , Imagen de Difusión Tensora/métodos , Sustancia Blanca/patología , Enfermedad de Parkinson/patología , Sustancia Negra/diagnóstico por imagen , Sustancia Negra/patología , Edema/patologíaRESUMEN
Background: Following chemoradiotherapy for high-grade glioma (HGG), it is often challenging to distinguish treatment changes from true tumor progression using conventional MRI. The diffusion basis spectrum imaging (DBSI) hindered fraction is associated with tissue edema or necrosis, which are common treatment-related changes. We hypothesized that DBSI hindered fraction may augment conventional imaging for earlier diagnosis of progression versus treatment effect. Methods: Adult patients were prospectively recruited if they had a known histologic diagnosis of HGG and completed standard-of-care chemoradiotherapy. DBSI and conventional MRI data were acquired longitudinally beginning 4 weeks post-radiation. Conventional MRI and DBSI metrics were compared with respect to their ability to diagnose progression versus treatment effect. Results: Twelve HGG patients were enrolled between August 2019 and February 2020, and 9 were ultimately analyzed (5 progression, 4 treatment effect). Within new or enlarging contrast-enhancing regions, DBSI hindered fraction was significantly higher in the treatment effect group compared to progression group (P = .0004). Compared to serial conventional MRI alone, inclusion of DBSI would have led to earlier diagnosis of either progression or treatment effect in 6 (66.7%) patients by a median of 7.7 (interquartile range = 0-20.1) weeks. Conclusions: In the first longitudinal prospective study of DBSI in adult HGG patients, we found that in new or enlarging contrast-enhancing regions following therapy, DBSI hindered fraction is elevated in cases of treatment effect compared to those with progression. Hindered fraction map may be a valuable adjunct to conventional MRI to distinguish tumor progression from treatment effect.
RESUMEN
BACKGROUND: Pretreatment identification of pathological extranodal extension (ENE) would guide therapy de-escalation strategies for in human papillomavirus (HPV)-associated oropharyngeal carcinoma but is diagnostically challenging. ECOG-ACRIN Cancer Research Group E3311 was a multicentre trial wherein patients with HPV-associated oropharyngeal carcinoma were treated surgically and assigned to a pathological risk-based adjuvant strategy of observation, radiation, or concurrent chemoradiation. Despite protocol exclusion of patients with overt radiographic ENE, more than 30% had pathological ENE and required postoperative chemoradiation. We aimed to evaluate a CT-based deep learning algorithm for prediction of ENE in E3311, a diagnostically challenging cohort wherein algorithm use would be impactful in guiding decision-making. METHODS: For this retrospective evaluation of deep learning algorithm performance, we obtained pretreatment CTs and corresponding surgical pathology reports from the multicentre, randomised de-escalation trial E3311. All enrolled patients on E3311 required pretreatment and diagnostic head and neck imaging; patients with radiographically overt ENE were excluded per study protocol. The lymph node with largest short-axis diameter and up to two additional nodes were segmented on each scan and annotated for ENE per pathology reports. Deep learning algorithm performance for ENE prediction was compared with four board-certified head and neck radiologists. The primary endpoint was the area under the curve (AUC) of the receiver operating characteristic. FINDINGS: From 178 collected scans, 313 nodes were annotated: 71 (23%) with ENE in general, 39 (13%) with ENE larger than 1 mm ENE. The deep learning algorithm AUC for ENE classification was 0·86 (95% CI 0·82-0·90), outperforming all readers (p<0·0001 for each). Among radiologists, there was high variability in specificity (43-86%) and sensitivity (45-96%) with poor inter-reader agreement (κ 0·32). Matching the algorithm specificity to that of the reader with highest AUC (R2, false positive rate 22%) yielded improved sensitivity to 75% (+ 13%). Setting the algorithm false positive rate to 30% yielded 90% sensitivity. The algorithm showed improved performance compared with radiologists for ENE larger than 1 mm (p<0·0001) and in nodes with short-axis diameter 1 cm or larger. INTERPRETATION: The deep learning algorithm outperformed experts in predicting pathological ENE on a challenging cohort of patients with HPV-associated oropharyngeal carcinoma from a randomised clinical trial. Deep learning algorithms should be evaluated prospectively as a treatment selection tool. FUNDING: ECOG-ACRIN Cancer Research Group and the National Cancer Institute of the US National Institutes of Health.
Asunto(s)
Carcinoma , Aprendizaje Profundo , Neoplasias Orofaríngeas , Infecciones por Papillomavirus , Humanos , Virus del Papiloma Humano , Estudios Retrospectivos , Infecciones por Papillomavirus/diagnóstico por imagen , Infecciones por Papillomavirus/complicaciones , Extensión Extranodal , Neoplasias Orofaríngeas/diagnóstico por imagen , Neoplasias Orofaríngeas/patología , Algoritmos , Carcinoma/complicaciones , Tomografía Computarizada por Rayos XRESUMEN
Purpose: Sarcopenia is an established prognostic factor in patients diagnosed with head and neck squamous cell carcinoma (HNSCC). The quantification of sarcopenia assessed by imaging is typically achieved through the skeletal muscle index (SMI), which can be derived from cervical neck skeletal muscle (SM) segmentation and cross-sectional area. However, manual SM segmentation is labor-intensive, prone to inter-observer variability, and impractical for large-scale clinical use. To overcome this challenge, we have developed and externally validated a fully-automated image-based deep learning (DL) platform for cervical vertebral SM segmentation and SMI calculation, and evaluated the relevance of this with survival and toxicity outcomes. Materials and Methods: 899 patients diagnosed as having HNSCC with CT scans from multiple institutes were included, with 335 cases utilized for training, 96 for validation, 48 for internal testing and 393 for external testing. Ground truth single-slice segmentations of SM at the C3 vertebra level were manually generated by experienced radiation oncologists. To develop an efficient method of segmenting the SM, a multi-stage DL pipeline was implemented, consisting of a 2D convolutional neural network (CNN) to select the middle slice of C3 section and a 2D U-Net to segment SM areas. The model performance was evaluated using the Dice Similarity Coefficient (DSC) as the primary metric for the internal test set, and for the external test set the quality of automated segmentation was assessed manually by two experienced radiation oncologists. The L3 skeletal muscle area (SMA) and SMI were then calculated from the C3 cross sectional area (CSA) of the auto-segmented SM. Finally, established SMI cut-offs were used to perform further analyses to assess the correlation with survival and toxicity endpoints in the external institution with univariable and multivariable Cox regression. Results: DSCs for validation set (n = 96) and internal test set (n = 48) were 0.90 (95% CI: 0.90 - 0.91) and 0.90 (95% CI: 0.89 - 0.91), respectively. The predicted CSA is highly correlated with the ground-truth CSA in both validation (r = 0.99, p < 0.0001) and test sets (r = 0.96, p < 0.0001). In the external test set (n = 377), 96.2% of the SM segmentations were deemed acceptable by consensus expert review. Predicted SMA and SMI values were highly correlated with the ground-truth values, with Pearson r ß 0.99 (p < 0.0001) for both the female and male patients in all datasets. Sarcopenia was associated with worse OS (HR 2.05 [95% CI 1.04 - 4.04], p = 0.04) and longer PEG tube duration (median 162 days vs. 134 days, HR 1.51 [95% CI 1.12 - 2.08], p = 0.006 in multivariate analysis. Conclusion: We developed and externally validated a fully-automated platform that strongly correlates with imaging-assessed sarcopenia in patients with H&N cancer that correlates with survival and toxicity outcomes. This study constitutes a significant stride towards the integration of sarcopenia assessment into decision-making for individuals diagnosed with HNSCC. SUMMARY STATEMENT: In this study, we developed and externally validated a deep learning model to investigate the impact of sarcopenia, defined as the loss of skeletal muscle mass, on patients with head and neck squamous cell carcinoma (HNSCC) undergoing radiotherapy. We demonstrated an efficient, fullyautomated deep learning pipeline that can accurately segment C3 skeletal muscle area, calculate cross-sectional area, and derive a skeletal muscle index to diagnose sarcopenia from a standard of care CT scan. In multi-institutional data, we found that pre-treatment sarcopenia was associated with significantly reduced overall survival and an increased risk of adverse events. Given the increased vulnerability of patients with HNSCC, the assessment of sarcopenia prior to radiotherapy may aid in informed treatment decision-making and serve as a predictive marker for the necessity of early supportive measures.
RESUMEN
BACKGROUND: Artificial intelligence (AI) and deep learning have shown great potential in streamlining clinical tasks. However, most studies remain confined to in silico validation in small internal cohorts, without external validation or data on real-world clinical utility. We developed a strategy for the clinical validation of deep learning models for segmenting primary non-small-cell lung cancer (NSCLC) tumours and involved lymph nodes in CT images, which is a time-intensive step in radiation treatment planning, with large variability among experts. METHODS: In this observational study, CT images and segmentations were collected from eight internal and external sources from the USA, the Netherlands, Canada, and China, with patients from the Maastro and Harvard-RT1 datasets used for model discovery (segmented by a single expert). Validation consisted of interobserver and intraobserver benchmarking, primary validation, functional validation, and end-user testing on the following datasets: multi-delineation, Harvard-RT1, Harvard-RT2, RTOG-0617, NSCLC-radiogenomics, Lung-PET-CT-Dx, RIDER, and thorax phantom. Primary validation consisted of stepwise testing on increasingly external datasets using measures of overlap including volumetric dice (VD) and surface dice (SD). Functional validation explored dosimetric effect, model failure modes, test-retest stability, and accuracy. End-user testing with eight experts assessed automated segmentations in a simulated clinical setting. FINDINGS: We included 2208 patients imaged between 2001 and 2015, with 787 patients used for model discovery and 1421 for model validation, including 28 patients for end-user testing. Models showed an improvement over the interobserver benchmark (multi-delineation dataset; VD 0·91 [IQR 0·83-0·92], p=0·0062; SD 0·86 [0·71-0·91], p=0·0005), and were within the intraobserver benchmark. For primary validation, AI performance on internal Harvard-RT1 data (segmented by the same expert who segmented the discovery data) was VD 0·83 (IQR 0·76-0·88) and SD 0·79 (0·68-0·88), within the interobserver benchmark. Performance on internal Harvard-RT2 data segmented by other experts was VD 0·70 (0·56-0·80) and SD 0·50 (0·34-0·71). Performance on RTOG-0617 clinical trial data was VD 0·71 (0·60-0·81) and SD 0·47 (0·35-0·59), with similar results on diagnostic radiology datasets NSCLC-radiogenomics and Lung-PET-CT-Dx. Despite these geometric overlap results, models yielded target volumes with equivalent radiation dose coverage to those of experts. We also found non-significant differences between de novo expert and AI-assisted segmentations. AI assistance led to a 65% reduction in segmentation time (5·4 min; p<0·0001) and a 32% reduction in interobserver variability (SD; p=0·013). INTERPRETATION: We present a clinical validation strategy for AI models. We found that in silico geometric segmentation metrics might not correlate with clinical utility of the models. Experts' segmentation style and preference might affect model performance. FUNDING: US National Institutes of Health and EU European Research Council.
Asunto(s)
Carcinoma de Pulmón de Células no Pequeñas , Aprendizaje Profundo , Neoplasias Pulmonares , Algoritmos , Inteligencia Artificial , Humanos , Tomografía Computarizada por Tomografía de Emisión de Positrones , Estados UnidosRESUMEN
Identifying the presence of intravenous contrast material on CT scans is an important component of data curation for medical imaging-based artificial intelligence model development and deployment. Use of intravenous contrast material is often poorly documented in imaging metadata, necessitating impractical manual annotation by clinician experts. Authors developed a convolutional neural network (CNN)-based deep learning platform to identify intravenous contrast enhancement on CT scans. For model development and validation, authors used six independent datasets of head and neck (HN) and chest CT scans, totaling 133 480 axial two-dimensional sections from 1979 scans, which were manually annotated by clinical experts. Five CNN models were trained first on HN scans for contrast enhancement detection. Model performances were evaluated at the patient level on a holdout set and external test set. Models were then fine-tuned on chest CT data and externally validated. This study found that Digital Imaging and Communications in Medicine metadata tags for intravenous contrast material were missing or erroneous for 1496 scans (75.6%). An EfficientNetB4-based model showed the best performance, with areas under the curve (AUCs) of 0.996 and 1.0 in HN holdout (n = 216) and external (n = 595) sets, respectively, and AUCs of 1.0 and 0.980 in the chest holdout (n = 53) and external (n = 402) sets, respectively. This automated, scan-to-prediction platform is highly accurate at CT contrast enhancement detection and may be helpful for artificial intelligence model development and clinical application. Keywords: CT, Head and Neck, Supervised Learning, Transfer Learning, Convolutional Neural Network (CNN), Machine Learning Algorithms, Contrast Material Supplemental material is available for this article. © RSNA, 2022.
RESUMEN
OBJECTIVE: To prospectively determine whether diffusion basis spectrum imaging (DBSI) detects, differentiates and quantitates coexisting inflammation, demyelination, axonal injury and axon loss in mice with optic neuritis (ON) due to experimental autoimmune encephalomyelitis (EAE), and to determine if DBSI accurately measures effects of fingolimod on underlying pathology. METHODS: EAE was induced in 7-week-old C57BL/6 female mice. Visual acuity (VA) was assessed daily to detect onset of ON after which daily oral-treatment with either fingolimod (1 mg/kg) or saline was given for ten weeks. In vivo DBSI scans of optic nerves were performed at baseline, 2-, 6- and 10-weeks post treatment. DBSI-derived metrics including restricted isotropic diffusion tensor fraction (putatively reflecting cellularity), non-restricted isotropic diffusion tensor fraction (putatively reflecting vasogenic edema), DBSI-derived axonal volume, axial diffusivity, λ⥠(putatively reflecting axonal integrity), and increased radial diffusivity, λ⥠(putatively reflecting demyelination). Mice were killed immediately after the last DBSI scan for immunohistochemical assessment. RESULTS: Optic nerves of fingolimod-treated mice exhibited significantly better (p < 0.05) VA than saline-treated group at each time point. During ten-week of treatment, DBSI-derived non-restricted and restricted-isotropic-diffusion-tensor fractions, and axonal volumes were not significantly different (p > 0.05) from the baseline values in fingolimod-treated mice. Transient DBSI-λ⥠decrease and DBSI-λ⥠increase were detected during Fingolimod treatment. DBSI-derived metrics assessed in vivo significantly correlated (p < 0.05) with the corresponding histological markers. CONCLUSION: DBSI was used to assess changes of the underlying optic nerve pathologies in EAE mice with ON, exhibiting great potential as a noninvasive outcome measure for monitoring disease progression and therapeutic efficacy for MS.