We consider the problem of a primary source tracking a moving object under time-varying and unknown noise conditions. We propose two methods that integrate sequential Bayesian filtering with transfer learning to improve tracking performance. Within the transfer learning framework, multiple sources are assumed to perform the same tracking task as the primary source but under different noise conditions. The first method uses Gaussian mixtures to model the measurement distribution, assuming that the measurement noise intensity at the learning sources is fixed and known a priori and the learning and primary sources are simultaneously tracking the same source. The second tracking method uses Dirichlet process mixtures to model noise parameters, assuming that the learning source measurement noise intensity is unknown. As we demonstrate, the use of Bayesian nonparametric learning does not require all sources to track the same object. The learned information can be stored and transferred to the primary source when needed. Using simulations for both high- and low-signal-to-noise ratio conditions, we demonstrate the improved primary tracking performance as the number of learning sources increases.

Background/Objectives: The parapharyngeal space is a complex anatomical region that houses critical neurovascular structures and serves as the origin of rare tumors, which account for 0.5–1% of head and neck neoplasms. Magnetic resonance imaging (MRI) is useful for their preoperative assessment. However, its accuracy in real-world clinical settings remains underexplored. This study aimed to investigate the diagnostic accuracy of MRI for parapharyngeal tumors at two tertiary centers. Methods: This retrospective study included patients who underwent MRI and surgical excision at two tertiary centers in Saudi Arabia between 2018 and 2024. Two reviewers independently extracted their MRI data and compared them with the final pathological data to determine the diagnostic performance of MRI. Results: Of the 31 patients (58.1% female; median age, 37.5 years), 90.3% had benign tumors. Neurogenic (41.9%) and salivary (25.8%) tumors were most common; 61.3% were located within the pre-styloid space. The benign and malignant groups had comparable baseline characteristics. MRI demonstrated moderate overall diagnostic agreement (κ = 0.525) and near-perfect concordance for schwannomas (κ = 0.912) and paragangliomas (κ = 0.839) but poor agreement for hemangiopericytomas (κ = −0.051). It had high accuracy (90.3%), specificity (92.9%), and negative predictive value (96.3%) for detecting malignancy but limited sensitivity (66.7%) or positive predictive value (50.0%). Nonetheless, cautious interpretation is required due to the limited prevalence of malignancy in the cohort (n = 3). Conclusions: MRI demonstrated high specificity for benign parapharyngeal space lesions in routine clinical reporting within this retrospective cohort, reflecting strong radiologic–pathologic agreement. Estimates of sensitivity and positive predictive value for malignancy were influenced by the limited number of malignant cases. Accordingly, the reported diagnostic performance measures should be interpreted as descriptive and exploratory, characterizing real-world MRI performance rather than definitive diagnostic accuracy.

We propose an algorithm that integrates Bayesian filtering with transfer learning to track a moving object under unknown time-varying environmental conditions. In order to account for measurement noise intensity variations in the primary source, we use multiple learning sources with labeled measurements. For each source, the measurement likelihood is modeled using Gaussian mixtures whose parameters are learned from conjugate priors. Weighted basis combinations of the multiple learned information are then used to model the measurement likelihood of the primary source; the basis weights are learned using a Dirichlet distribution prior. The improved tracking performance of the proposed algorithm is demonstrated for both low and high noise scenarios.