Dataset Information

Multiparametric Quantitative Imaging Biomarkers for Phenotype Classification: A Framework for Development and Validation.

ABSTRACT: This manuscript is the third in a five-part series related to statistical assessment methodology for technical performance of multi-parametric quantitative imaging biomarkers (mp-QIBs). We outline approaches and statistical methodologies for developing and evaluating a phenotype classification model from a set of multiparametric QIBs. We then describe validation studies of the classifier for precision, diagnostic accuracy, and interchangeability with a comparator classifier. We follow with an end-to-end real-world example of development and validation of a classifier for atherosclerotic plaque phenotypes. We consider diagnostic accuracy and interchangeability to be clinically meaningful claims for a phenotype classification model informed by mp-QIB inputs, aiming to provide tools to demonstrate agreement between imaging-derived characteristics and clinically established phenotypes. Understanding that we are working in an evolving field, we close our manuscript with an acknowledgement of existing challenges and a discussion of where additional work is needed. In particular, we discuss the challenges involved with technical performance and analytical validation of mp-QIBs. We intend for this manuscript to further advance the robust and promising science of multiparametric biomarker development.

SUBMITTER: Delfino JG

PROVIDER: S-EPMC9825632 | biostudies-literature | 2023 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Multiparametric Quantitative Imaging Biomarkers for Phenotype Classification: A Framework for Development and Validation.

Delfino Jana G JG Pennello Gene A GA Barnhart Huiman X HX Buckler Andrew J AJ Wang Xiaofeng X Huang Erich P EP Raunig Dave L DL Guimaraes Alexander R AR Hall Timothy J TJ deSouza Nandita M NM Obuchowski Nancy N

Academic radiology 20221004 2

This manuscript is the third in a five-part series related to statistical assessment methodology for technical performance of multi-parametric quantitative imaging biomarkers (mp-QIBs). We outline approaches and statistical methodologies for developing and evaluating a phenotype classification model from a set of multiparametric QIBs. We then describe validation studies of the classifier for precision, diagnostic accuracy, and interchangeability with a comparator classifier. We follow with an en ...[more]

PMID: 36202670

Similar Datasets

Project description:While it has been established that a number of microenvironment components can affect the likelihood of metastasis, the link between microenvironment and tumor cell phenotypes is poorly understood. Here we have examined microenvironment control over two different tumor cell motility phenotypes required for metastasis. By high-resolution multiphoton microscopy of mammary carcinoma in mice, we detected two phenotypes of motile tumor cells, different in locomotion speed. Only slower tumor cells exhibited protrusions with molecular, morphological, and functional characteristics associated with invadopodia. Each region in the primary tumor exhibited either fast- or slow-locomotion. To understand how the tumor microenvironment controls invadopodium formation and tumor cell locomotion, we systematically analyzed components of the microenvironment previously associated with cell invasion and migration. No single microenvironmental property was able to predict the locations of tumor cell phenotypes in the tumor if used in isolation or combined linearly. To solve this, we utilized the support vector machine (SVM) algorithm to classify phenotypes in a nonlinear fashion. This approach identified conditions that promoted either motility phenotype. We then demonstrated that varying one of the conditions may change tumor cell behavior only in a context-dependent manner. In addition, to establish the link between phenotypes and cell fates, we photoconverted and monitored the fate of tumor cells in different microenvironments, finding that only tumor cells in the invadopodium-rich microenvironments degraded extracellular matrix (ECM) and disseminated. The number of invadopodia positively correlated with degradation, while the inhibiting metalloproteases eliminated degradation and lung metastasis, consistent with a direct link among invadopodia, ECM degradation, and metastasis. We have detected and characterized two phenotypes of motile tumor cells in vivo, which occurred in spatially distinct microenvironments of primary tumors. We show how machine-learning analysis can classify heterogeneous microenvironments in vivo to enable prediction of motility phenotypes and tumor cell fate. The ability to predict the locations of tumor cell behavior leading to metastasis in breast cancer models may lead towards understanding the heterogeneity of response to treatment.

Project description:Alzheimer's disease is the primary cause of dementia worldwide, with an increasing morbidity burden that may outstrip diagnosis and management capacity as the population ages. Current methods integrate patient history, neuropsychological testing and MRI to identify likely cases, yet effective practices remain variably applied and lacking in sensitivity and specificity. Here we report an interpretable deep learning strategy that delineates unique Alzheimer's disease signatures from multimodal inputs of MRI, age, gender, and Mini-Mental State Examination score. Our framework linked a fully convolutional network, which constructs high resolution maps of disease probability from local brain structure to a multilayer perceptron and generates precise, intuitive visualization of individual Alzheimer's disease risk en route to accurate diagnosis. The model was trained using clinically diagnosed Alzheimer's disease and cognitively normal subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset (n = 417) and validated on three independent cohorts: the Australian Imaging, Biomarker and Lifestyle Flagship Study of Ageing (AIBL) (n = 382), the Framingham Heart Study (n = 102), and the National Alzheimer's Coordinating Center (NACC) (n = 582). Performance of the model that used the multimodal inputs was consistent across datasets, with mean area under curve values of 0.996, 0.974, 0.876 and 0.954 for the ADNI study, AIBL, Framingham Heart Study and NACC datasets, respectively. Moreover, our approach exceeded the diagnostic performance of a multi-institutional team of practicing neurologists (n = 11), and high-risk cerebral regions predicted by the model closely tracked post-mortem histopathological findings. This framework provides a clinically adaptable strategy for using routinely available imaging techniques such as MRI to generate nuanced neuroimaging signatures for Alzheimer's disease diagnosis, as well as a generalizable approach for linking deep learning to pathophysiological processes in human disease.

Project description:BackgroundAtaxia Telangiectasia (A-T) is an inherited multisystem disorder with cerebellar neurodegeneration. The relationships between imaging metrics of cerebellar health and neurological function across childhood in A-T are unknown, but may be important for determining timing and impact of therapeutic interventions.PurposeTo test the hypothesis that abnormalities of cerebellar structure, physiology and cellular health occur in childhood A-T and correlate with neurological disability, we performed multiparametric cerebellar MRI and establish associations with disease status in childhood A-T.MethodsProspective cross-sectional observational study. 22 young people (9 females / 13 males, age 6.6-17.8 years) with A-T and 24 matched healthy controls underwent 3-Tesla MRI with volumetric, diffusion and proton spectroscopic acquisitions. Participants with A-T underwent structured neurological assessment, and expression / activity of ataxia-telangiectasia mutated (ATM) kinase were recorded.ResultsAtaxia-telangiectasia participants had cerebellar volume loss (fractional total cerebellar volume: 5.3% vs 8.7%, P < 0.0005, fractional 4th ventricular volumes: 0.19% vs 0.13%, P < 0.0005), that progressed with age (fractional cerebellar volumes, r = -0.66, P = 0.001), different from the control group (t = -4.88, P < 0.0005). The relationship between cerebellar volume and age was similar for A-T participants with absent ATM kinase production and those producing non-functioning ATM kinase. Markers of cerebellar white matter injury were elevated in ataxia-telangiectasia vs controls (apparent diffusion coefficient: 0.89 × 10-3 mm2 s-1 vs 0.69 × 10-3 mm2 s-1, p < 0.0005) and correlated (age-corrected) with neurometabolite ratios indicating impaired neuronal viability (N-acetylaspartate:creatine r = -0.70, P < 0.001); gliosis (inositol:creatine r = 0.50, P = 0.018; combined glutamine/glutamate:creatine r = -0.55, P = 0.008) and increased myelin turnover (choline:creatine r = 0.68, P < 0.001). Fractional 4th ventricular volume was the only variable retained in the regression model predicting neurological function (adjusted r2 = 0.29, P = 0.015).ConclusionsQuantitative MRI demonstrates cerebellar abnormalities in children with A-T, providing non-invasive measures of progressive cerebellar injury and markers reflecting neurological status. These MRI metrics may be of value in determining timing and impact of interventions aimed at altering the natural history of A-T.

Project description:ObjectivesTo develop and validate a multiparametric model to predict neoadjuvant treatment response in rectal cancer at baseline using a heterogeneous multicenter MRI dataset.MethodsBaseline staging MRIs (T2W (T2-weighted)-MRI, diffusion-weighted imaging (DWI) / apparent diffusion coefficient (ADC)) of 509 patients (9 centres) treated with neoadjuvant chemoradiotherapy (CRT) were collected. Response was defined as (1) complete versus incomplete response, or (2) good (Mandard tumor regression grade (TRG) 1-2) versus poor response (TRG3-5). Prediction models were developed using combinations of the following variable groups: (1) Non-imaging: age/sex/tumor-location/tumor-morphology/CRT-surgery interval (2) Basic staging: cT-stage/cN-stage/mesorectal fascia involvement, derived from (2a) original staging reports, or (2b) expert re-evaluation (3) Advanced staging: variables from 2b combined with cTN-substaging/invasion depth/extramural vascular invasion/tumor length (4) Quantitative imaging: tumour volume + first-order histogram features (from T2W-MRI and DWI/ADC) Models were developed with data from 6 centers (n = 412) using logistic regression with the Least Absolute Shrinkage and Selector Operator (LASSO) feature selection, internally validated using repeated (n = 100) random hold-out validation, and externally validated using data from 3 centers (n = 97).ResultsAfter external validation, the best model (including non-imaging and advanced staging variables) achieved an area under the curve of 0.60 (95%CI=0.48-0.72) to predict complete response and 0.65 (95%CI=0.53-0.76) to predict a good response. Quantitative variables did not improve model performance. Basic staging variables consistently achieved lower performance compared to advanced staging variables.ConclusionsOverall model performance was moderate. Best results were obtained using advanced staging variables, highlighting the importance of good-quality staging according to current guidelines. Quantitative imaging features had no added value (in this heterogeneous dataset).Clinical relevance statementPredicting tumour response at baseline could aid in tailoring neoadjuvant therapies for rectal cancer. This study shows that image-based prediction models are promising, though are negatively affected by variations in staging quality and MRI acquisition, urging the need for harmonization.Key pointsThis multicenter study combining clinical information and features derived from MRI rendered disappointing performance to predict response to neoadjuvant treatment in rectal cancer. Best results were obtained with the combination of clinical baseline information and state-of-the-art image-based staging variables, highlighting the importance of good quality staging according to current guidelines and staging templates. No added value was found for quantitative imaging features in this multicenter retrospective study. This is likely related to acquisition variations, which is a major problem for feature reproducibility and thus model generalizability.

Project description:PurposeTo develop and independently validate prognostic imaging biomarkers for predicting survival in patients with glioblastoma on the basis of multiregion quantitative image analysis.Materials and methodsThis retrospective study was approved by the local institutional review board, and informed consent was waived. A total of 79 patients from two independent cohorts were included. The discovery and validation cohorts consisted of 46 and 33 patients with glioblastoma from the Cancer Imaging Archive (TCIA) and the local institution, respectively. Preoperative T1-weighted contrast material-enhanced and T2-weighted fluid-attenuation inversion recovery magnetic resonance (MR) images were analyzed. For each patient, we semiautomatically delineated the tumor and performed automated intratumor segmentation, dividing the tumor into spatially distinct subregions that demonstrate coherent intensity patterns across multiparametric MR imaging. Within each subregion and for the entire tumor, we extracted quantitative imaging features, including those that fully capture the differential contrast of multimodality MR imaging. A multivariate sparse Cox regression model was trained by using TCIA data and tested on the validation cohort.ResultsThe optimal prognostic model identified five imaging biomarkers that quantified tumor surface area and intensity distributions of the tumor and its subregions. In the validation cohort, our prognostic model achieved a concordance index of 0.67 and significant stratification of overall survival by using the log-rank test (P = .018), which outperformed conventional prognostic factors, such as age (concordance index, 0.57; P = .389) and tumor volume (concordance index, 0.59; P = .409).ConclusionThe multiregion analysis presented here establishes a general strategy to effectively characterize intratumor heterogeneity manifested at multimodality imaging and has the potential to reveal useful prognostic imaging biomarkers in glioblastoma.

Dataset Information

Multiparametric Quantitative Imaging Biomarkers for Phenotype Classification: A Framework for Development and Validation.

Publications

Multiparametric Quantitative Imaging Biomarkers for Phenotype Classification: A Framework for Development and Validation.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets