Background: Artificial intelligence (AI) is reshaping neuroimaging workflows for Alzheimer’s disease (AD) diagnosis, particularly through PET and MRI analysis advances. Since the FDA approval of Tauvid, a PET tracer targeting tau pathology, there has been a notable increase in studies applying AI to neuroimaging data. This narrative review synthesizes recent, high-impact literature to highlight clinically relevant AI applications in AD imaging. Methods: This review examined peer-reviewed studies published between January 2020 and January 2025, focusing on the use of AI, including machine learning, deep learning, and hybrid models for diagnostic and prognostic tasks in AD using PET and/or MRI. Studies were identified through targeted PubMed, Scopus, and Embase searches, emphasizing methodological diversity and clinical relevance. Results: A total of 109 studies were categorized into five thematic areas: Image preprocessing and segmentation, diagnostic classification, prognosis and disease staging, multimodal data fusion, and emerging innovations. Deep learning models such as convolutional neural networks (CNNs), generative adversarial networks (GANs), and transformer-based architectures were widely employed by the research community in the field of AD. At the same time, several models reported strong diagnostic performance, but methodological challenges such as reproducibility, small sample sizes, and lack of external validation limit clinical translation. Trends in explainable AI, synthetic imaging, and integration of clinical biomarkers are also discussed. Conclusions: AI is rapidly advancing the field of AD imaging, offering tools for enhanced segmentation, staging, and early diagnosis. Multimodal approaches and biomarker-guided models show particular promise. However, future research must focus on reproducibility, interpretability, and standardized validation to bridge the gap between research and clinical practice.

Background: ^99mTc-macroaggregated albumin (^99mTc-MAA) scintigraphy is utilized in treatment planning for Yttrium-90 (⁹⁰Y) Selective Internal Radiation Therapy (SIRT) of liver tumors to evaluate hepatopulmonary shunting by calculating the lung shunt fraction (LSF). The purpose of this study was to evaluate if LSF calculation using SPECT/CT instead of planar gamma camera imaging is more accurate and if this can potentially lead to more effective treatment planning of hepatic lesions while avoiding excessive pulmonary irradiation. Results: LSF calculation was obtained using two different methodologies in 85 cases from consecutive patients intended to receive ⁹⁰Y SIRT. The first method was based on planar gamma camera imaging in the anterior and posterior views with geometric mean calculation of the LSF from regions of interest of the liver and lungs. The second method was based on segmentation of the liver and lungs from SPECT/CT images of the thorax and abdomen. The differences in planar imaging versus SPECT/CT derived LSF values along with the estimated absorbed lung mean dose (LMD) were evaluated. The LSF values were higher in planar imaging versus SPECT/CT in 81/85 cases, with a mean value of 8.5% vs. 4.6% respectively; the difference was statistically significant using a paired t-test (alpha = 0.05). In those patients who received SIRT, the estimated absorbed LMD calculated with planar imaging was significantly higher than with SPECT/CT (t-test, P < 0.005). Repeated phantom experiments using an anthropomorphic torso phantom with variable ^99mTc activity concentrations for the liver and lungs were performed with the standard patient protocol, demonstrated improved accuracy of the LSF calculation based on SPECT/CT than planar imaging (mean overestimated value of 6% vs. 26%). Conclusions: This study demonstrates that LSF calculation using planar imaging can be significantly overestimated while calculation using SPECT/CT imaging and appropriate segmentation tools can be more accurate. Minimizing the errors in obtaining the LSF can lead to more effective ⁹⁰Y SIRT treatment planning for hepatic tumors while ensuring the lung dose will not exceed the standard acceptable safety thresholds.

In transarterial radioembolization (TARE) of hepatocellular carcinoma (HCC) with Yttrium-90 (Y-90) microspheres, recent studies correlate dosimetry from bremsstrahlung single photon emission tomography (SPECT/CT) with treatment outcomes; however, these studies focus on measures of central tendency rather than volumetric coverage metrics commonly used in radiation oncology. We hypothesized that three-dimensional (3D) isodose coverage of gross tumor volume (GTV) is the driving factor in HCC treatment response to TARE and is best assessed using advanced dosimetry techniques applied to nuclear imaging of actual Y-90 biodistribution. We reviewed 51 lobar TARE Y-90 treatments of 43 HCC patients. Dose prescriptions were 120 Gy for TheraSpheres and 85 Gy for SIR-Spheres. All patients underwent post-TARE Y-90 bremsstrahlung SPECT/CT imaging. Commercial software was used to contour gross tumor volume (GTV) and liver on post-TARE SPECT/CT. Y-90 dose distributions were calculated using the Local Deposition Model based on post-TARE SPECT/CT activity maps. Median gross tumor volume (GTV) dose; GTV receiving less than 100 Gy, 70 Gy and 50 Gy; minimum dose covering the hottest 70%, 95%, and 98% of the GTV (D70, D95, D98); mean dose to nontumorous liver, and disease burden (GTV/liver volume) were obtained. Clinical outcomes were collected for all patients by chart and imaging review. HCC treatment response was assessed according to the modified response criteria in solid tumors (mRECIST) guidelines. Kaplan-Meier (KM) survival estimates and multivariate regression analyses (MVA) were performed using STATA. Median survival was 22.5 months for patients achieving objective response (OR) in targeted lesions (complete response (CR) or partial response (PR) per mRECIST) vs. 7.6 months for non-responders (NR, stable disease or disease progression per mRECIST). On MVA, the volume of underdosed tumor (GTV receiving less than 100 Gy) was the only significant dosimetric predictor for CR (p = 0.0004) and overall survival (OS, p = 0.003). All targets with less than CR (n = 39) had more than 20 cc of underdosed tumor. D70 (p = 0.038) correlated with OR, with mean D70 of 95 Gy for responders and 60 Gy for non-responders (p = 0.042). On MVA, mean dose to nontumorous liver trended toward significant association with grade 3+ toxicity (p = 0.09) and correlated with delivered activity (p < 0.001) and burden of disease (p = 0.05). Dosimetric models supplied area under the curve estimates of > 0.80 predicting CR, OR, and ≥grade 3 acute toxicity. Dosimetric parameters derived from the retrospective analysis of post-TARE Y-90 bremsstrahlung SPECT/CT after lobar treatment of HCC suggest that volumetric coverage of GTV, not a high mean or median dose, is the driving factor in treatment response and that this is best assessed through the analysis of actual Y-90 biodistribution.

Background: This narrative review summarizes recent progress in artificial intelligence (AI), especially radiomics and deep learning, for non-invasive diagnosis and molecular profiling of gliomas. Methodology: A thorough literature search was conducted on PubMed, Scopus, and Embase for studies published from January 2020 to July 2025, focusing on clinical and technical research. In key areas, these studies examine AI models’ predictive capabilities with multi-parametric Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET). Results: The domains identified in the literature include the advancement of radiomic models for tumor grading and biomarker prediction, such as Isocitrate Dehydrogenase (IDH) mutation, O6-methylguanine-dna methyltransferase (MGMT) promoter methylation, and 1p/19q codeletion. The growing use of convolutional neural networks (CNNs) and generative adversarial networks (GANs) in tumor segmentation, classification, and prognosis was also a significant topic discussed in the literature. Deep learning (DL) methods are evaluated against traditional radiomics regarding feature extraction, scalability, and robustness to imaging protocol differences across institutions. Conclusions: This review analyzes emerging efforts to combine clinical, imaging, and histology data within hybrid or transformer-based AI systems to enhance diagnostic accuracy. Significant findings include the application of DL to predict cyclin-dependent kinase inhibitor 2A/B (CDKN2A/B) deletion and chemokine CCL2 expression. These highlight the expanding capabilities of imaging-based genomic inference and the importance of clinical data in multimodal fusion. Challenges such as data harmonization, model interpretability, and external validation still need to be addressed.

Thyroid cancer, specifically differentiated thyroid carcinoma (DTC), is one of the most prevalent endocrine malignancies worldwide. Radioactive iodine therapy (RAIT) using I-131 has been a standard-of-care approach for DTC due to its ability to ablate remnant thyroid disease following surgery, thus reducing the risk of recurrence. It is also used for the treatment of iodine-avid metastases. RAIT dosimetry can be employed to determine the optimal treatment dose of I-131 to effectively treat cancer cells while safeguarding against undesirable radiation effects such as bone marrow toxicity or radiation pneumonitis. Conventional dosimetry protocols for RAIT, however, are complex and time-consuming, involving multiple days of imaging and blood sampling. This study explores the use of Artificial Intelligence (AI) in simplifying and optimizing RAIT. A retrospective analysis was conducted on 83 adult patients with DTC who underwent RAIT dosimetry at our institution between 1996 and 2023. The conventional MIRD-based dosimetry protocol involved imaging and blood sampling at 4, 24, 48, 72, and 96 h post-administration of a tracer activity of I-131. An AI system based on a deep-learning neural network was developed to predict the maximum permissible activity (MPA) for RAIT using only the data obtained from the initial 4, 24, and 48 h time points. The AI system predicted the MPA values with high accuracy, showing no significant difference compared to the results obtained from conventional MIRD-based analysis utilizing a paired t-test (p = 0.351, 95% CI). The developed AI system offers the potential to streamline the dosimetry process, reducing the number of imaging and blood sampling sessions while also optimizing resource allocation. Additionally, the AI approach can uncover underlying relationships in data that were previously unknown. Our findings suggest that AI-based dosimetry may be a promising method for patient-specific treatment planning in differentiated thyroid carcinoma, representing a step towards applying precision medicine for thyroid cancer. Further validation and implementation studies are warranted to assess the clinical applicability of the AI system.

Introduction: MGMT promoter methylation is a critical predictive biomarker in high-grade gliomas (HGG), but its assessment currently relies on invasive tissue sampling. We aimed to develop an explainable, modality-adaptive, and calibrated radiomics model for non-invasive prediction of MGMT promoter methylation using multi-center MRI data. Methods: Pre-operative MRI from the UCSF-PDGM and UPENN-GBM cohorts was analyzed using radiomics extracted from intratumoral and peritumoral regions. Conventional (T1, T2, FLAIR) and advanced (DWI/ADC, ASL) MRI sequences were included. A novel modality-adaptive framework was implemented, allowing the model to automatically ignore advanced modalities when unavailable. After feature ranking and redundancy reduction, six machine-learning classifiers were optimized and Platt-calibrated. Model performance was evaluated on a held-out test set using ROC metrics, calibration assessment, and Decision Curve Analysis (DCA). Feature contributions were interpreted using SHAP. Results: The top-performing LightGBM model, trained on the 500 most important radiomic features, achieved an AUC of 0.67, recall of 0.90, and accuracy of 0.72 on the independent test set. The model demonstrated strong sensitivity for identifying methylated tumors, minimizing false-negative predictions. Calibration improved clinical net benefit across a range of threshold probabilities on DCA. Feature attribution analysis revealed balanced contributions from conventional and advanced MRI modalities, with texture and intensity-based descriptors being most influential. Notably, low-frequency FLAIR wavelet intensity features were associated with unmethylated tumors. Conclusion: This explainable, modality-adaptive radiomics model identified biologically consistent MGMT-related imaging patterns and demonstrated decision-analytic value for clinical risk stratification. The framework supports real-world applicability in heterogeneous imaging environments. Future work should focus on external validation and integration with clinical and molecular biomarkers to further enhance predictive performance.