An official website of the United States government
The burgeoning field of brain health research increasingly leverages artificial intelligence (AI) to analyze and interpret neuroimaging data. Medical foundation models have shown promise of superior performance with better sample efficiency. This work introduces a novel approach towards creating 3-dimensional (3D) medical foundation models for multimodal neuroimage segmentation through self-supervised training. Our approach involves a novel two-stage pretraining approach using vision transformers. The first stage encodes anatomical structures in generally healthy brains from the large-scale unlabeled neuroimage dataset of multimodal brain magnetic resonance imaging (MRI) images from 41,400 participants. This stage of pertaining focuses on identifying key features such as shapes and sizes of different brain structures. The second pretraining stage identifies disease-specific attributes, such as geometric shapes of tumors and lesions and spatial placements within the brain. This dual-phase methodology significantly reduces the extensive data requirements usually necessary for AI model training in neuroimage segmentation with the flexibility to adapt to various imaging modalities. We rigorously evaluate our model, BrainSegFounder, using the Brain Tumor Segmentation (BraTS) challenge and Anatomical Tracings of Lesions After Stroke v2.0 (ATLAS v2.0) datasets. BrainSegFounder demonstrates a significant performance gain, surpassing the achievements of the previous winning solutions using fully supervised learning. Our findings underscore the impact of scaling up both the model complexity and the volume of unlabeled training data derived from generally healthy brains. Both of these factors enhance the accuracy and predictive capabilities of the model in neuroimage segmentation tasks. Our pretrained models and code are at https://github.com/lab-smile/BrainSegFounder.
more »
« less
Assessment of brain cancer atlas maps with multimodal imaging features
Abstract Background Glioblastoma Multiforme (GBM) is a fast-growing and highly aggressive brain tumor that invades the nearby brain tissue and presents secondary nodular lesions across the whole brain but generally does not spread to distant organs. Without treatment, GBM can result in death in about 6 months. The challenges are known to depend on multiple factors: brain localization, resistance to conventional therapy, disrupted tumor blood supply inhibiting effective drug delivery, complications from peritumoral edema, intracranial hypertension, seizures, and neurotoxicity. Main text Imaging techniques are routinely used to obtain accurate detections of lesions that localize brain tumors. Especially magnetic resonance imaging (MRI) delivers multimodal images both before and after the administration of contrast, which results in displaying enhancement and describing physiological features as hemodynamic processes. This review considers one possible extension of the use of radiomics in GBM studies, one that recalibrates the analysis of targeted segmentations to the whole organ scale. After identifying critical areas of research, the focus is on illustrating the potential utility of an integrated approach with multimodal imaging, radiomic data processing and brain atlases as the main components. The templates associated with the outcome of straightforward analyses represent promising inference tools able to spatio-temporally inform on the GBM evolution while being generalizable also to other cancers. Conclusions The focus on novel inference strategies applicable to complex cancer systems and based on building radiomic models from multimodal imaging data can be well supported by machine learning and other computational tools potentially able to translate suitably processed information into more accurate patient stratifications and evaluations of treatment efficacy. Graphical Abstract
more »
« less
- Award ID(s):
- 2313443
- PAR ID:
- 10435811
- Date Published:
- Journal Name:
- Journal of Translational Medicine
- Volume:
- 21
- Issue:
- 1
- ISSN:
- 1479-5876
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract BackgroundDelta radiomics is a high‐throughput computational technique used to describe quantitative changes in serial, time‐series imaging by considering the relative change in radiomic features of images extracted at two distinct time points. Recent work has demonstrated a lack of prognostic signal of radiomic features extracted using this technique. We hypothesize that this lack of signal is due to the fundamental assumptions made when extracting features via delta radiomics, and that other methods should be investigated. PurposeThe purpose of this work was to show a proof‐of‐concept of a new radiomics paradigm for sparse, time‐series imaging data, where features are extracted from a spatial‐temporal manifold modeling the time evolution between images, and to assess the prognostic value on patients with oropharyngeal cancer (OPC). MethodsTo accomplish this, we developed an algorithm to mathematically describe the relationship between two images acquired at time and . These images serve as boundary conditions of a partial differential equation describing the transition from one image to the other. To solve this equation, we propagate the position and momentum of each voxel according to Fokker–Planck dynamics (i.e., a technique common in statistical mechanics). This transformation is driven by an underlying potential force uniquely determined by the equilibrium image. The solution generates a spatial‐temporal manifold (3 spatial dimensions + time) from which we define dynamic radiomic features. First, our approach was numerically verified by stochastically sampling dynamic Gaussian processes of monotonically decreasing noise. The transformation from high to low noise was compared between our Fokker–Planck estimation and simulated ground‐truth. To demonstrate feasibility and clinical impact, we applied our approach to18F‐FDG‐PET images to estimate early metabolic response of patients (n = 57) undergoing definitive (chemo)radiation for OPC. Images were acquired pre‐treatment and 2‐weeks intra‐treatment (after 20 Gy). Dynamic radiomic features capturing changes in texture and morphology were then extracted. Patients were partitioned into two groups based on similar dynamic radiomic feature expression via k‐means clustering and compared by Kaplan–Meier analyses with log‐rank tests (p < 0.05). These results were compared to conventional delta radiomics to test the added value of our approach. ResultsNumerical results confirmed our technique can recover image noise characteristics given sparse input data as boundary conditions. Our technique was able to model tumor shrinkage and metabolic response. While no delta radiomics features proved prognostic, Kaplan–Meier analyses identified nine significant dynamic radiomic features. The most significant feature was Gray‐Level‐Size‐Zone‐Matrix gray‐level variance (p = 0.011), which demonstrated prognostic improvement over its corresponding delta radiomic feature (p = 0.722). ConclusionsWe developed, verified, and demonstrated the prognostic value of a novel, physics‐based radiomics approach over conventional delta radiomics via data assimilation of quantitative imaging and differential equations.more » « less
-
Abstract BackgroundGlioblastoma Multiforme, an aggressive primary brain tumor, has a poor prognosis and no effective standard of care treatments. Most patients undergoing radiotherapy, along with Temozolomide chemotherapy, develop resistance to the drug, and recurrence of the tumor is a common issue after the treatment. We propose to model the pathways active in Glioblastoma using Boolean network techniques. The network captures the genetic interactions and possible mutations that are involved in the development of the brain tumor. The model is used to predict the theoretical efficacies of drugs for the treatment of cancer. ResultsWe use the Boolean network to rank the critical intervention points in the pathway to predict an effective therapeutic strategy for Glioblastoma. Drug repurposing helps to identify non-cancer drugs that could be effective in cancer treatment. We predict the effectiveness of drug combinations of anti-cancer and non-cancer drugs for Glioblastoma. ConclusionsGiven the genetic profile of a GBM tumor, the Boolean model can predict the most effective targets for treatment. We also identified two-drug combinations that could be more effective in killing GBM cells than conventional chemotherapeutic agents. The non-cancer drug Aspirin could potentially increase the cytotoxicity of TMZ in GBM patients.more » « less
-
null (Ed.)Glioblastoma (GBM) is one of the most aggressive forms of adult brain cancers and is highly resistant to treatment, with a median survival of 12–18 months after diagnosis. The poor survival is due to its infiltrative pattern of invasion into the normal brain parenchyma, the diffuse nature of its growth, and its ability to quickly grow, spread, and relapse. Temozolomide is a well-known FDA-approved alkylating chemotherapy agent used for the treatment of high-grade malignant gliomas, and it has been shown to improve overall survival. However, in most cases, the tumor relapses. In recent years, CAP has been used as an emerging technology for cancer therapy. The purpose of this study was to implement a combination therapy of CAP and TMZ to enhance the effect of TMZ and apparently sensitize GBMs. In vitro evaluations in TMZ-sensitive and resistant GBM cell lines established a CAP chemotherapy enhancement and potential sensitization effect across various ranges of CAP jet application. This was further supported with in vivo findings demonstrating that a single CAP jet applied non-invasively through the skull potentially sensitizes GBM to subsequent treatment with TMZ. Gene functional enrichment analysis further demonstrated that co-treatment with CAP and TMZ resulted in a downregulation of cell cycle pathway genes. These observations indicate that CAP can be potentially useful in sensitizing GBM to chemotherapy and for the treatment of glioblastoma as a non-invasive translational therapy.more » « less
-
Abstract Glioblastoma (GBM) is hard to treat due to cellular invasion into functioning brain tissues, limited drug delivery, and evolved treatment resistance. Recurrence is nearly universal even after surgery, chemotherapy, and radiation. Photodynamic therapy (PDT) involves photosensitizer administration followed by light activation to generate reactive oxygen species at tumor sites, thereby killing cells or inducing biological changes. PDT can ablate unresectable GBM and sensitize tumors to chemotherapy. Verteporfin (VP) is a promising photosensitizer that relies on liposomal carriers for clinical use. While lipids increase VP's solubility, they also reduce intracellular photosensitizer accumulation. Here, a pure‐drug nanoformulation of VP, termed “NanoVP”, eliminating the need for lipids, excipients, or stabilizers is reported. NanoVP has a tunable size (65–150 nm) and 1500‐fold higher photosensitizer loading capacity than liposomal VP. NanoVP shows a 2‐fold increase in photosensitizer uptake and superior PDT efficacy in GBM cells compared to liposomal VP. In mouse models, NanoVP‐PDT improved tumor control and extended animal survival, outperforming liposomal VP and 5‐aminolevulinic acid (5‐ALA). Moreover, low‐dose NanoVP‐PDT can safely open the blood‐brain barrier, increasing drug accumulation in rat brains by 5.5‐fold compared to 5‐ALA. NanoVP is a new photosensitizer formulation that has the potential to facilitate PDT for the treatment of GBM.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1186/s12967-023-04222-3
