In search of truth: Evaluating concordance of AI-based anatomy segmentation models

Purpose AI-based methods for anatomy segmentation can help automate characterization of large imaging datasets. The growing number of similar in functionality models raises the challenge of evaluating them on datasets that do not contain ground truth annotations. We introduce a practical framework to assist in this task. Approach We harmonize the segmentation results into a standard, interoperable representation, which enables consistent, terminology-based labeling of the structures. We extend 3D Slicer to streamline loading and comparison of these harmonized segmentations, and demonstrate how standard representation simplifies review of the results using interactive summary plots and browser-based visualization using OHIF Viewer. To demonstrate the utility of the approach we apply it to evaluating segmentation of 31 anatomical structures (lungs, vertebrae, ribs, and heart) by six open-source models - TotalSegmentator 1.5 and 2.6, Auto3DSeg, MOOSE, MultiTalent, and CADS - for a sample of Computed Tomography (CT) scans from the publicly available National Lung Screening Trial (NLST) dataset. Results We demonstrate the utility of the framework in enabling automating loading, structure-wise inspection and comparison across models. Preliminary results ascertain practical utility of the approach in allowing quick detection and review of problematic results. The comparison shows excellent agreement segmenting some (e.g., lung) but not all structures (e.g., some models produce invalid vertebrae or rib segmentations). Conclusions The resources developed are linked from https://imagingdatacommons.github.io/segmentation-comparison/ including segmentation harmonization scripts, summary plots, and visualization tools. This work assists in model evaluation in absence of ground truth, ultimately enabling informed model selection.

From Data to Diagnosis: A Large, Comprehensive Bone Marrow Dataset and AI Methods for Childhood Leukemia Prediction

Leukemia diagnosis primarily relies on manual microscopic analysis of bone marrow morphology supported by additional laboratory parameters, making it complex and time consuming. While artificial intelligence (AI) solutions have been proposed, most utilize private datasets and only cover parts of the diagnostic pipeline. Therefore, we present a large, high-quality, publicly available leukemia bone marrow dataset spanning the entire diagnostic process, from cell detection to diagnosis. Using this dataset, we further propose methods for cell detection, cell classification, and diagnosis prediction. The dataset comprises 246 pediatric patients with diagnostic, clinical and laboratory information, over 40 000 cells with bounding box annotations and more than 28 000 of these with high-quality class labels, making it the most comprehensive dataset publicly available. Evaluation of the AI models yielded an average precision of 0.96 for the cell detection, an area under the curve of 0.98, and an F1-score of 0.61 for the 33-class cell classification, and a mean F1-score of 0.90 for the diagnosis prediction using predicted cell counts. While the proposed approaches demonstrate their usefulness for AI-assisted diagnostics, the dataset will foster further research and development in the field, ultimately contributing to more precise diagnoses and improved patient outcomes.

Benchmarking of Deep Learning Methods for Generic MRI Multi-OrganAbdominal Segmentation

Recent advances in deep learning have led to robust automated tools for segmentation of abdominal computed tomography (CT). Meanwhile, segmentation of magnetic resonance imaging (MRI) is substantially more challenging due to the inherent signal variability and the increased effort required for annotating training datasets. Hence, existing approaches are trained on limited sets of MRI sequences, which might limit their generalizability. To characterize the landscape of MRI abdominal segmentation tools, we present here a comprehensive benchmarking of the three state-of-the-art and open-source models: MRSegmentator, MRISegmentator-Abdomen, and TotalSegmentator MRI. Since these models are trained using labor-intensive manual annotation cycles, we also introduce and evaluate ABDSynth, a SynthSeg-based model purely trained on widely available CT segmentations (no real images). More generally, we assess accuracy and generalizability by leveraging three public datasets (not seen by any of the evaluated methods during their training), which span all major manufacturers, five MRI sequences, as well as a variety of subject conditions, voxel resolutions, and fields-of-view. Our results reveal that MRSegmentator achieves the best performance and is most generalizable. In contrast, ABDSynth yields slightly less accurate results, but its relaxed requirements in training data make it an alternative when the annotation budget is limited. The evaluation code and datasets are given for future benchmarking at https://github.com/deepakri201/AbdoBench, along with inference code and weights for ABDSynth.

Rule-based outlier detection of AI-generated anatomy segmentations

There is a dire need for medical imaging datasets with accompanying annotations to perform downstream patient analysis. However, it is difficult to manually generate these annotations, due to the time-consuming nature, and the variability in clinical conventions. Artificial intelligence has been adopted in the field as a potential method to annotate these large datasets, however, a lack of expert annotations or ground truth can inhibit the adoption of these annotations. We recently made a dataset publicly available including annotations and extracted features of up to 104 organs for the National Lung Screening Trial using the TotalSegmentator method. However, the released dataset does not include expert-derived annotations or an assessment of the accuracy of the segmentations, limiting its usefulness. We propose the development of heuristics to assess the quality of the segmentations, providing methods to measure the consistency of the annotations and a comparison of results to the literature. We make our code and related materials publicly available at https://github.com/ImagingDataCommons/CloudSegmentatorResults and interactive tools at https://huggingface.co/spaces/ImagingDataCommons/CloudSegmentatorResults.

Automatic classification of prostate MR series type using image content and metadata

With the wealth of medical image data, efficient curation is essential. Assigning the sequence type to magnetic resonance images is necessary for scientific studies and artificial intelligence-based analysis. However, incomplete or missing metadata prevents effective automation. We therefore propose a deep-learning method for classification of prostate cancer scanning sequences based on a combination of image data and DICOM metadata. We demonstrate superior results compared to metadata or image data alone, and make our code publicly available at https://github.com/deepakri201/DICOMScanClassification.

Towards Automatic Abdominal MRI Organ Segmentation: Leveraging Synthesized Data Generated From CT Labels

Deep learning has shown great promise in the ability to automatically annotate organs in magnetic resonance imaging (MRI) scans, for example, of the brain. However, despite advancements in the field, the ability to accurately segment abdominal organs remains difficult across MR. In part, this may be explained by the much greater variability in image appearance and severely limited availability of training labels. The inherent nature of computed tomography (CT) scans makes it easier to annotate, resulting in a larger availability of expert annotations for the latter. We leverage a modality-agnostic domain randomization approach, utilizing CT label maps to generate synthetic images on-the-fly during training, further used to train a U-Net segmentation network for abdominal organs segmentation. Our approach shows comparable results compared to fully-supervised segmentation methods trained on MR data. Our method results in Dice scores of 0.90 (0.08) and 0.91 (0.08) for the right and left kidney respectively, compared to a pretrained nnU-Net model yielding 0.87 (0.20) and 0.91 (0.03). We will make our code publicly available.

Real-Time Dynamic Data Driven Deformable Registration for Image-Guided Neurosurgery: Computational Aspects

Current neurosurgical procedures utilize medical images of various modalities to enable the precise location of tumors and critical brain structures to plan accurate brain tumor resection. The difficulty of using preoperative images during the surgery is caused by the intra-operative deformation of the brain tissue (brain shift), which introduces discrepancies concerning the preoperative configuration. Intra-operative imaging allows tracking such deformations but cannot fully substitute for the quality of the pre-operative data. Dynamic Data Driven Deformable Non-Rigid Registration (D4NRR) is a complex and time-consuming image processing operation that allows the dynamic adjustment of the pre-operative image data to account for intra-operative brain shift during the surgery. This paper summarizes the computational aspects of a specific adaptive numerical approximation method and its variations for registering brain MRIs. It outlines its evolution over the last 15 years and identifies new directions for the computational aspects of the technique.

Enrichment of the NLST and NSCLC-Radiomics computed tomography collections with AI-derived annotations

Public imaging datasets are critical for the development and evaluation of automated tools in cancer imaging. Unfortunately, many do not include annotations or image-derived features, complicating their downstream analysis. Artificial intelligence-based annotation tools have been shown to achieve acceptable performance and thus can be used to automatically annotate large datasets. As part of the effort to enrich public data available within NCI Imaging Data Commons (IDC), here we introduce AI-generated annotations for two collections of computed tomography images of the chest, NSCLC-Radiomics, and the National Lung Screening Trial. Using publicly available AI algorithms we derived volumetric annotations of thoracic organs at risk, their corresponding radiomics features, and slice-level annotations of anatomical landmarks and regions. The resulting annotations are publicly available within IDC, where the DICOM format is used to harmonize the data and achieve FAIR principles. The annotations are accompanied by cloud-enabled notebooks demonstrating their use. This study reinforces the need for large, publicly accessible curated datasets and demonstrates how AI can be used to aid in cancer imaging.

The NCI Imaging Data Commons as a platform for reproducible research in computational pathology

Objective: Reproducibility is critical for translating machine learning-based (ML) solutions in computational pathology (CompPath) into practice. However, an increasing number of studies report difficulties in reproducing ML results. The NCI Imaging Data Commons (IDC) is a public repository of >120 cancer image collections, including >38,000 whole-slide images (WSIs), that is designed to be used with cloud-based ML services. Here, we explore the potential of the IDC to facilitate reproducibility of CompPath research. Materials and Methods: The IDC realizes the FAIR principles: All images are encoded according to the DICOM standard, persistently identified, discoverable via rich metadata, and accessible via open tools. Taking advantage of this, we implemented two experiments in which a representative ML-based method for classifying lung tumor tissue was trained and/or evaluated on different datasets from the IDC. To assess reproducibility, the experiments were run multiple times with independent but identically configured sessions of common ML services. Results: The AUC values of different runs of the same experiment were generally consistent and in the same order of magnitude as a similar, previously published study. However, there were occasional small variations in AUC values of up to 0.044, indicating a practical limit to reproducibility. Discussion and conclusion: By realizing the FAIR principles, the IDC enables other researchers to reuse exactly the same datasets. Cloud-based ML services enable others to run CompPath experiments in an identically configured computing environment without having to own high-performance hardware. The combination of both makes it possible to approach the reproducibility limit.

Repeatability of Multiparametric Prostate MRI Radiomics Features

In this study we assessed the repeatability of the values of radiomics features for small prostate tumors using test-retest? Multiparametric Magnetic Resonance Imaging (mpMRI) images. The premise of radiomics is that quantitative image features can serve as biomarkers characterizing disease. For such biomarkers to be useful, repeatability is a basic requirement, meaning its value must remain stable between two scans, if the conditions remain stable. We investigated repeatability of radiomics features under various preprocessing and extraction configurations including various image normalization schemes, different image pre-filtering, 2D vs 3D texture computation, and different bin widths for image discretization. Image registration as means to re-identify regions of interest across time points was evaluated against human-expert segmented regions in both time points. Even though we found many radiomics features and preprocessing combinations with a high repeatability (Intraclass Correlation Coefficient (ICC) > 0.85), our results indicate that overall the repeatability is highly sensitive to the processing parameters (under certain configurations, it can be below 0.0). Image normalization, using a variety of approaches considered, did not result in consistent improvements in repeatability. There was also no consistent improvement of repeatability through the use of pre-filtering options, or by using image registration between timepoints to improve consistency of the region of interest localization. Based on these results we urge caution when interpreting radiomics features and advise paying close attention to the processing configuration details of reported results. Furthermore, we advocate reporting all processing details in radiomics studies and strongly recommend making the implementation available.