Objective:To develop radiomics score (RS) based on 18F-FDG PET/CT, and construct the machine learning model combining clinical and other relevant factors for personalized prediction of 2-year event-free survival (2-EFS) in patients with diffuse large B-cell lymphoma (DLBCL), and to perform interpretability analysis of the model. Methods:A total of 91 patients (49 males, 42 females; age (57.8±12.8) years) with pathologically confirmed DLBCL from December 2017 to December 2020 at the First Hospital of Shanxi Medical University were retrospectively analyzed. According to the ratio of 7∶3, patients were randomly divided into training set ( n=63) and test set ( n=28), and divided into non-progression group and progression group according to the follow-up results. The whole-body PET semi-quantitative parameters were calculated from the PET/CT images before treatment, and 328 radiomics features were extracted from the largest target lesions of patients. The least absolute shrinkage and selection operator (LASSO) was used to develop the RS. Clinical and PET characteristic difference analysis was performed through χ2 test and Mann-Whitney U test. Extreme gradient boosting (XGBoost) models were constructed based on clinical, PET radiomics features and RS, and the prediction efficiency of each model was evaluated by ROC AUC. The model interpretability was analyzed by using shapely additive explanation (SHAP). Results:Of all patients, 32 had disease progression and 59 did not. There were no significant differences in baseline characteristics between the training set and the test set ( χ2 values: 0.06-1.84, U values: 665.00-763.00, all P>0.05). The comparison between the progression group and non-progression group in the training set showed statistical differences in the international prognostic index (IPI) score ( χ2=4.87, P=0.027), myelocytomatosis viral oncogene (MYC) protein expression ( χ2=4.29, P=0.038), and metabolic tumor volume (MTV; U=307.00, P=0.038). Seven radiomics features were screened by LASSO. Among XGBoost models with different feature combinations, IPI score, MYC protein expression, MTV combined with RS had the highest predictive efficiency (training set: AUC=0.73; test set: AUC=0.70). Through SHAP analysis, RS was the most predictive feature in the optimal model. Conclusion:The machine learning integrated model of IPI score, MYC protein expression and MTV combined with RS can effectively predict the prognosis of DLBCL patients, and baseline 18F-FDG PET/CT radiomics can be used as a potential means to evaluate the prognosis of DLBCL patients.

In recent years, immunotherapy has become another important treatment for non-small cell lung cancer (NSCLC) after surgery, radiotherapy, chemotherapy and targeted therapy. As an emerging non-invasive tool, PET/CT radiomics provides a new way for precise diagnosis and treatment of NSCLC by quantitatively analyzing the metabolic and structural characteristics of tumors, demonstrating significant promise for applications in NSCLC immunotherapy. This article aims to review the research progress of PET/CT radiomics in NSCLC immunotherapy and to explore its potential value in predicting biomarkers, therapeutic efficacy, adverse effects, and evaluating prognosis.

Objective:To develop radiomics score (RS) based on 18F-FDG PET/CT, and construct the machine learning model combining clinical and other relevant factors for personalized prediction of 2-year event-free survival (2-EFS) in patients with diffuse large B-cell lymphoma (DLBCL), and to perform interpretability analysis of the model. Methods:A total of 91 patients (49 males, 42 females; age (57.8±12.8) years) with pathologically confirmed DLBCL from December 2017 to December 2020 at the First Hospital of Shanxi Medical University were retrospectively analyzed. According to the ratio of 7∶3, patients were randomly divided into training set ( n=63) and test set ( n=28), and divided into non-progression group and progression group according to the follow-up results. The whole-body PET semi-quantitative parameters were calculated from the PET/CT images before treatment, and 328 radiomics features were extracted from the largest target lesions of patients. The least absolute shrinkage and selection operator (LASSO) was used to develop the RS. Clinical and PET characteristic difference analysis was performed through χ2 test and Mann-Whitney U test. Extreme gradient boosting (XGBoost) models were constructed based on clinical, PET radiomics features and RS, and the prediction efficiency of each model was evaluated by ROC AUC. The model interpretability was analyzed by using shapely additive explanation (SHAP). Results:Of all patients, 32 had disease progression and 59 did not. There were no significant differences in baseline characteristics between the training set and the test set ( χ2 values: 0.06-1.84, U values: 665.00-763.00, all P>0.05). The comparison between the progression group and non-progression group in the training set showed statistical differences in the international prognostic index (IPI) score ( χ2=4.87, P=0.027), myelocytomatosis viral oncogene (MYC) protein expression ( χ2=4.29, P=0.038), and metabolic tumor volume (MTV; U=307.00, P=0.038). Seven radiomics features were screened by LASSO. Among XGBoost models with different feature combinations, IPI score, MYC protein expression, MTV combined with RS had the highest predictive efficiency (training set: AUC=0.73; test set: AUC=0.70). Through SHAP analysis, RS was the most predictive feature in the optimal model. Conclusion:The machine learning integrated model of IPI score, MYC protein expression and MTV combined with RS can effectively predict the prognosis of DLBCL patients, and baseline 18F-FDG PET/CT radiomics can be used as a potential means to evaluate the prognosis of DLBCL patients.

In recent years, immunotherapy has become another important treatment for non-small cell lung cancer (NSCLC) after surgery, radiotherapy, chemotherapy and targeted therapy. As an emerging non-invasive tool, PET/CT radiomics provides a new way for precise diagnosis and treatment of NSCLC by quantitatively analyzing the metabolic and structural characteristics of tumors, demonstrating significant promise for applications in NSCLC immunotherapy. This article aims to review the research progress of PET/CT radiomics in NSCLC immunotherapy and to explore its potential value in predicting biomarkers, therapeutic efficacy, adverse effects, and evaluating prognosis.

Radiation-induced heart disease (RIHD) is a common complication of radiotherapy and one of the main causes of non-cancer death in patients with thoracic malignant tumors, which seriously affects the clinical outcome and quality of life of patients. PET imaging is able to assess myocardial perfusion, abnormal myocardial metabolic, cardiac sympathetic disorders, myocardial fibrosis and inflammation, and is of great value in the early diagnosis and detection of RIHD. This article reviews the progress of PET imaging in the early detection of RIHD.

Objective:To prepare nanoprobes by using polydopamine (PDA) as a carrier which is modified with the sonosensitizer protoporphyrin Ⅸ (PpⅨ) and labeled with 131I, 99Tc m or 177Lu, and to explore the value of these new nanoprobes in diagnosis and combination therapy of breast cancer. Methods:PDA particles were synthesized by aqueous oxidation, and a layer of polyethylene glycol (PEG) and PpⅨ were modified on the surface to product PDA-PEG-PpⅨ. Then the nuclides 131I, 99Tc m and 177Lu were labeled on PDA, respectively, and the labeling yield and stability were determined. The cytotoxicity test was conducted by comparing the viabilities of 4T1 tumor cells in free 131I group and 131I-PDA-PEG-PpⅨ group. The 4T1 cells were divided into 7 groups according to different treatment methods: PDA-PEG-PpⅨ group, PDA-PEG-PpⅨ+ photothermal therapy (PTT) group, PDA-PEG-PpⅨ+ sonodynamic therapy (SDT) group, 131I-PDA-PEG-PpⅨ+ PTT group, 131I-PDA-PEG-PpⅨ+ SDT group, 131I-PDA-PEG-PpⅨ+ PTT+ SDT group (100 μg/ml PDA-PEG-PpⅨ, 925 kBq/ml 131I), and the control group (DMEM culture medium). The cell viabilities of those groups were compared to evaluate the therapeutic effect. 4T1 tumor bearing mouse models were established, then 99Tc m-PDA-PEG-PpⅨ was injected through the tail vein (29.6 MBq) or intratumorally (14.8 MBq) to perform gamma imaging. The independent-sample t test and one-way analysis of variance were used for data analysis. Results:The PDA particles were uniform in size, with a particle size of (160.0±1.5) nm. They had a good photothermal conversion effect. A characteristic peak consistent with PpⅨ (400 nm) appeared in the UV-Vis absorption spectrum of PDA-PEG-PpIX. In the cytotoxicity test, when the radioactivity was 1.850 or 3.700 or 7.400 MBq/ml, the cell viabilities of free 131I group and 131I-PDA-PEG-PpⅨ group were significantly different ((72.18±6.57)% vs (86.07±5.17)%, (59.31±9.06)% vs (80.85±4.21)%, (42.90±1.30)% vs (72.99±5.73)%; t values: 3.71, 4.82, 11.46, P values: 0.006, 0.001, <0.001). The 131I-PDA-PEG-PpⅨ+ PTT+ SDT combination therapy had a better killing effect on 4T1 tumor cells than the combination of 131I-PDA-PEG-PpⅨ+ PTT and 131I-PDA-PEG-PpⅨ+ SDT (cell viabilities: (10.09±2.50)% vs (16.04±2.63)%, (28.65±4.72)%; F=351.66, P<0.001). In vivo imaging showed that 99Tc m-PDA-PEG-PpⅨ was stable in mouse models and could be effectively enriched in tumors. Conclusions:A multifunctional nanoprobe based on PDA is successfully prepared. The radionuclide labeling method is simple and effective, with a good stability. 131I-PDA-PEG-PpⅨ can kill 4T1 cells efficiently. 99Tc m-PDA-PEG-PpⅨ has an obvious tumor concentration effect in mouse models.