Abstract: In certain aspects, a method includes analyzing a single 4D flow magnetic resonance imaging (MRI) scan associated with a blood vessel of a patient, wherein the MRI scan is taken at an initial time point. The method includes determining a critical threshold beyond which an area of the blood vessel in the MRI scan fluctuates unboundedly under infinitesimal perturbations. The method includes selecting a base flow comprising a periodic limit cycle following a pulsatile waveform of blood pressure over a cardiac cycle associated with the blood vessel in the MRI scan. The method includes generating a flutter parameter for describing an onset of an instability triggering fluttering of a vessel wall of the blood vessel in the MRI scan, wherein generating the flutter parameter is based on at least the analyzing of the single 4D flow MRI scan, determining the critical threshold, and the selecting the base flow.

Abstract: A first Vision-Language Model (VLM) in a first branch identifies a first set of fields in input data using a visualized first set of bounding boxes (BB). The first VLM labels and outputs a labeled first set of fields. A first agentic AI in the first branch localizes and outputs an identified field as a desired type of field using a visualized identified BB. A second VLM in a second branch identifies a second set of fields in the input data using a visualized second set of BBs. An MLLM uses the input data with the second set of BBs to output a set of recognizing field from the second set of BBs. A second agentic AI in the second branch and labels a target field. A training data set is formed by combining the input data, labeled identified field, and the labeled target field.

Abstract: Systems and methods include a diagnostic evaluation environment that may provide probabilities of occurrence for one or more conditions based on evaluation of an input dataset. The probabilities may be based on multivariate evaluation of the input dataset against one or more pre-determined clinical metrics. In operation, a distributed environment may process the input dataset to visualize information for a given range and then provide probabilities and associated metric values for different occurrences and conditions to provide diagnostic support to treat one or more conditions.

Abstract: Esophageal bolus transport and esophageal mechanics are quantified from medical imaging data, such as dynamic magnetic resonance imaging (“MRI”) data, computed tomography (“CT”) data, or the like. A machine learning model is used to process geometric or other spatiotemporal parameters of a bolus imaged with medical imaging in order to estimate quantitative parameters of the bolus and/or esophagus, such as cross-sectional area, fluid velocity, and fluid pressure. From these values, other parameters can be computed, such as esophageal stiffness and active relaxation.

Abstract: Input data including a facsimile representation of a paper document is input at first branch of a zero-shot configuration (Zcon), which includes a first Vision-Language Model (VLM) adapted to identify a first set of fields in the input data using a visualized first set of bounding boxes (BB). A first agentic AI in the first branch localizes a field as a desired type of filed using a visualized identified BB. A second VLM in a second branch identifies a second set of fields in the input data, and outputs a visualized second set of BBs. The input data with the second set of BBs is passed to a MLLM executing in the second branch, which outputs a set of recognizing fields within the BBs of the second set of BBs. A second agentic AI in the second branch localizes at least one recognized field as a target field.

Abstract: A first Vision-Language Model (VLM) in a first branch identifies a first set of fields in input data using a visualized first set of bounding boxes (BB). The first VLM labels and outputs a labeled first set of fields. A first agentic AI in the first branch localizes and outputs an identified field as a desired type of filed using a visualized identified BB. A second VLM in a second branch identifies a second set of fields in the input data using a visualized second set of BBs. An MLLM uses the input data with the second set of BBs to output a set of recognizing field from the second set of BBs. A second agentic AI in the second branch and labels a target field. A training data set is formed by combining the input data, labeled identified field, and the labeled target field.

Abstract: Flow through tubular organs (e.g., the esophagus) is analyzed based on fluid mechanics analysis of medical images. Using computational fluid dynamics, a reduced-order model is constructed and implemented to predict flow rate and fluid pressure developed inside flexible tubular organs inside the body. As one non-limiting example, the constructed model can be applied to analyze esophageal transport using fluoroscopy image sequences to predict flow rate, pressure, esophagus wall stiffness, and active relaxation.

Abstract: Esophageal bolus transport and esophageal mechanics are quantified from medical imaging data, such as dynamic magnetic resonance imaging (“MRI”) data, computed tomography (“CT”) data, or the like. A machine learning model is used to process geometric or other spatiotemporal parameters of a bolus imaged with medical imaging in order to estimate quantitative parameters of the bolus and/or esophagus, such as cross-sectional area, fluid velocity, and fluid pressure. From these values, other parameters can be computed, such as esophageal stiffness and active relaxation.

Abstract: A fluid mechanics-based analysis tool is implemented to compute pressure field data, fluid velocity data, and/or muscular work data in a lumen or other tubular organ or structure from planimetry and pressure data (e.g., measured using a balloon dilation or other planimetry catheter). In this way, flow data can be estimated, which are otherwise insensible from current planimetry catheter technologies due to economic and/or manufacturing limitations.