Generalizability and quality control of deep learning-based 2D echocardiography segmentation models in a large clinical dataset.

Use of machine learning (ML) for automated annotation of heart structures from echocardiographic videos is an active research area, but understanding of comparative, generalizable performance among models is lacking. This study aimed to (1) assess the generalizability of five state-of-the-art ML-based echocardiography segmentation models within a large Geisinger clinical dataset, and (2) test the hypothesis that a quality control (QC) method based on segmentation uncertainty can further improve segmentation results. Five models were applied to 47,431 echocardiography studies that were independent from any training samples. Chamber volume and mass from model segmentations were compared to clinically-reported values. The median absolute errors (MAE) in left ventricular (LV) volumes and ejection fraction exhibited by all five models were comparable to reported inter-observer errors (IOE). MAE for left atrial volume and LV mass were similarly favorable to respective IOE for models trained for those tasks. A single model consistently exhibited the lowest MAE in all five clinically-reported measures. We leveraged the tenfold cross-validation training scheme of this best-performing model to quantify segmentation uncertainty. We observed that removing segmentations with high uncertainty from 14 to 71% studies reduced volume/mass MAE by 6-10%. The addition of convexity filters improved specificity, efficiently removing < 10% studies with large MAE (16-40%). In conclusion, five previously published echocardiography segmentation models generalized to a large, independent clinical dataset-segmenting one or multiple cardiac structures with overall accuracy comparable to manual analyses-with variable performance. Convexity-reinforced uncertainty QC efficiently improved segmentation performance and may further facilitate the translation of such models.

A Machine Learning Approach to Management of Heart Failure Populations.

BACKGROUND: Heart failure is a prevalent, costly disease for which new value-based payment models demand optimized population management strategies. OBJECTIVES: This study sought to generate a strategy for managing populations of patients with heart failure by leveraging large clinical datasets and machine learning. METHODS: Geisinger electronic health record data were used to train machine learning models to predict 1-year all-cause mortality in 26,971 patients with heart failure who underwent 276,819 clinical episodes. There were 26 clinical variables (demographics, laboratory test results, medications), 90 diagnostic codes, 41 electrocardiogram measurements and patterns, 44 echocardiographic measurements, and 8 evidence-based "care gaps": flu vaccine, blood pressure of <130/80 mm Hg, A1c of <8%, cardiac resynchronization therapy, and active medications (active angiotensin-converting enzyme inhibitor/angiotensin II receptor blocker/angiotensin receptor-neprilysin inhibitor, aldosterone receptor antagonist, hydralazine, and evidence-based beta-blocker) were collected. Care gaps represented actionable variables for which associations with all-cause mortality were modeled from retrospective data and then used to predict the benefit of prospective interventions in 13,238 currently living patients. RESULTS: Machine learning models achieved areas under the receiver-operating characteristic curve (AUCs) of 0.74 to 0.77 in a split-by-year training/test scheme, with the nonlinear XGBoost model (AUC: 0.77) outperforming linear logistic regression (AUC: 0.74). Out of 13,238 currently living patients, 2,844 were predicted to die within a year, and closing all care gaps was predicted to save 231 of these lives. Prioritizing patients for intervention by using the predicted reduction in 1-year mortality risk outperformed all other priority rankings (e.g., random selection or Seattle Heart Failure risk score). CONCLUSIONS: Machine learning can be used to priority-rank patients most likely to benefit from interventions to optimize evidence-based therapies. This approach may prove useful for optimizing heart failure population health management teams within value-based payment models.

Thalamus segmentation using multi-modal feature classification: Validation and pilot study of an age-matched cohort.

Automatic segmentation of the thalamus can be used to measure differences and track changes in thalamic volume that may occur due to disease, injury or normal aging. An automatic thalamus segmentation algorithm incorporating features from diffusion tensor imaging (DTI) and thalamus priors constructed from multiple atlases is proposed. Multiple atlases with corresponding manual thalamus segmentations are registered to the target image and averaged to generate the thalamus prior. At each voxel in a region of interest around the thalamus, a multidimensional feature vector that includes the thalamus prior as well as a set of DTI features, including fractional anisotropy, mean diffusivity, and fiber orientation is formed. A random forest is trained to classify each voxel as belonging to the thalamus or background within the region of interest. Using a leave-one-out cross-validation on nine subjects, the proposed algorithm achieves a mean Dice score of 0.878 and 0.890 for the left and right thalami, respectively, which are higher Dice scores than the three state-of-art methods we compared to. We demonstrate the utility of the method with a pilot study exploring the difference in the thalamus fraction between 21 multiple sclerosis (MS) patients and 21 age-matched healthy controls. The left and right thalamic volumes (normalized by intracranial volumes) are larger in healthy controls by 7.6% and 7.3% respectively, compared to MS patients (though neither result is statistically significant).

Thalamus parcellation using multi-modal feature classification and thalamic nuclei priors.

Segmentation of the thalamus and thalamic nuclei is useful to quantify volumetric changes from neurodegenerative diseases. Most thalamus segmentation algorithms only use T1-weighted magnetic resonance images and current thalamic parcellation methods require manual interaction. Smaller nuclei, such as the lateral and medial geniculates, are challenging to locate due to their small size. We propose an automated segmentation algorithm using a set of features derived from diffusion tensor image (DTI) and thalamic nuclei location priors. After extracting features, a hierarchical random forest classifier is trained to locate the thalamus. A second random forest classifies thalamus voxels as belonging to one of six thalamic nuclei classes. The proposed algorithm was tested using a leave-one-out cross validation scheme and compared with state-of-the-art algorithms. The proposed algorithm has a higher Dice score compared to other methods for the whole thalamus and several nuclei.

Automatic method for thalamus parcellation using multi-modal feature classification.

Segmentation and parcellation of the thalamus is an important step in providing volumetric assessment of the impact of disease n brain structures. Conventionally, segmentation is carried out on T1-weighted magnetic resonance (MR) images and nuclear parcellation using diffusion weighted MR images. We present the first fully automatic method that incorporates both tissue contrasts and several derived fea-fractional anisotrophy, fiber orientation from the 5D Knutsson representation of the principal eigenvectors, and connectivity between the thalamus and the cortical lobes, as features. Combining these multiple information sources allows us to identify discriminating dimensions and thus parcellate the thalamic nuclei. A hierarchical random forest framework with a multidimensional feature per voxel, first distinguishes thalamus from background, and then separates each group of thalamic nuclei. Using a leave one out cross-validation on 12 subjects we have a mean Dice score of 0.805 and 0.799 for the left and right thalami, respectively. We also report overlap for the thalamic nuclear groups.

THALAMIC PARCELLATION FROM MULTI-MODAL DATA USING RANDOM FOREST LEARNING.

The thalamus sub-cortical gray matter structure consists of contiguous nuclei, each individually responsible for communication between various cerebral cortex and midbrain regions. These nuclei are differentially affected in neurodegenerative diseases such as multiple sclerosis and Alzheimer's. However thalamic parcellation of the nuclei, manual or automatic, is difficult given the limited contrast in any particular magnetic resonance (MR) modality. Several groups have had qualitative success differentiating nuclei based on spatial location and fiber orientation information in diffusion tensor imaging (DTI). In this paper, we extend these principles by combining these discriminating dimensions with structural MR and derived information, and by building random forest learners on the resultant multi-modal features. In training, we form a multi-dimensional feature per voxel, which we associate with a nucleus classification from a manual rater. Learners are trained to differentiate thalamus from background and thalamic nuclei from other nuclei. These learners inform the external forces of a multiple object level set model. Our cross-validated quantitative results on a set of twenty subjects show the efficacy and reproducibility of our results.

Regional appearance in deformable model segmentation.

Automated medical image segmentation is a challenging task that benefits from the use of effective image appearance models. In this paper, we compare appearance models at three regional scales for statistically characterizing image intensity near object boundaries in the context of segmentation via deformable models. The three models capture appearance in the form of regional intensity quantile functions. These distribution-based regional image descriptors are amenable to Euclidean methods such as principal component analysis, which we use to build the statistical appearance models. The first model uses two regions, the interior and exterior of the organ of interest. The second model accounts for exterior inhomogeneity by clustering on object-relative local intensity quantile functions to determine tissue-consistent regions relative to the organ boundary. The third model analyzes these image descriptors per geometrically defined local region. To evaluate the three models, we present segmentation results on bladders and prostates in CT in the context of day-to-day adaptive radiotherapy for the treatment of prostate cancer. Results show improved segmentations with more local regions, probably because smaller regions better represent local inhomogeneity in the intensity distribution near the organ boundary.