Deep learning model integrating positron emission tomography and clinical data for prognosis prediction in non-small cell lung cancer patients.

BACKGROUND: Lung cancer is the leading cause of cancer-related deaths worldwide. The majority of lung cancers are non-small cell lung cancer (NSCLC), accounting for approximately 85% of all lung cancer types. The Cox proportional hazards model (CPH), which is the standard method for survival analysis, has several limitations. The purpose of our study was to improve survival prediction in patients with NSCLC by incorporating prognostic information from F-18 fluorodeoxyglucose positron emission tomography (FDG PET) images into a traditional survival prediction model using clinical data. RESULTS: The multimodal deep learning model showed the best performance, with a C-index and mean absolute error of 0.756 and 399 days under a five-fold cross-validation, respectively, followed by ResNet3D for PET (0.749 and 405 days) and CPH for clinical data (0.747 and 583 days). CONCLUSION: The proposed deep learning-based integrative model combining the two modalities improved the survival prediction in patients with NSCLC.

Ultrasensitive detection of malignant melanoma using PET molecular imaging probes.

Malignant melanoma has one of the highest mortality rates of any cancer because of its aggressive nature and high metastatic potential. Clinical staging of the disease at the time of diagnosis is very important for the prognosis and outcome of melanoma treatment. In this study, we designed and synthesized the 18F-labeled pyridine-based benzamide derivatives N-(2-(dimethylamino)ethyl)-5-[18F]fluoropicolinamide ([18F]DMPY2) and N-(2-(dimethylamino)ethyl)-6-[18F]fluoronicotinamide ([18F]DMPY3) to detect primary and metastatic melanoma at an early stage and evaluated their performance in this task. [18F]DMPY2 and [18F]DMPY3 were synthesized by direct radiofluorination of the bromo precursor, and radiochemical yields were ∼15-20%. Cell uptakes of [18F]DMPY2 and [18F]DMPY3 were >103-fold and 18-fold higher, respectively, in B16F10 (mouse melanoma) cells than in negative control cells. Biodistribution studies revealed strong tumor uptake and retention of [18F]DMPY2 (24.8% injected dose per gram of tissue [ID/g] at 60 min) and [18F]DMPY3 (11.7%ID/g at 60 min) in B16F10 xenografts. MicroPET imaging of both agents demonstrated strong tumoral uptake/retention and rapid washout, resulting in excellent tumor-to-background contrast in B16F10 xenografts. In particular, [18F]DMPY2 clearly visualized almost all metastatic lesions in lung and lymph nodes, with excellent image quality. [18F]DMPY2 demonstrated a significantly higher tumor-to-liver ratio than [18F]fluorodeoxyglucose ([18F]FDG) and the previously reported benzamide tracers N-[2-(diethylamino)-ethyl]-5-[18F]fluoropicolinamide ([18F]P3BZA) and N-[2-(diethylamino)-ethyl]-4-[18F]fluorobenzamide ([18F]FBZA) in B16F10-bearing or SK-MEL-3 (human melanoma)-bearing mice. In conclusion, [18F]DMPY2 might have strong potential for the diagnosis of early stage primary and metastatic melanoma using positron emission tomography (PET).

Survival time prediction by integrating cox proportional hazards network and distribution function network.

BACKGROUND: The Cox proportional hazards model is commonly used to predict hazard ratio, which is the risk or probability of occurrence of an event of interest. However, the Cox proportional hazard model cannot directly generate an individual survival time. To do this, the survival analysis in the Cox model converts the hazard ratio to survival times through distributions such as the exponential, Weibull, Gompertz or log-normal distributions. In other words, to generate the survival time, the Cox model has to select a specific distribution over time. RESULTS: This study presents a method to predict the survival time by integrating hazard network and a distribution function network. The Cox proportional hazards network is adapted in DeepSurv for the prediction of the hazard ratio and a distribution function network applied to generate the survival time. To evaluate the performance of the proposed method, a new evaluation metric that calculates the intersection over union between the predicted curve and ground truth was proposed. To further understand significant prognostic factors, we use the 1D gradient-weighted class activation mapping method to highlight the network activations as a heat map visualization over an input data. The performance of the proposed method was experimentally verified and the results compared to other existing methods. CONCLUSIONS: Our results confirmed that the combination of the two networks, Cox proportional hazards network and distribution function network, can effectively generate accurate survival time.

Esophagus Segmentation in CT Images via Spatial Attention Network and STAPLE Algorithm.

One essential step in radiotherapy treatment planning is the organ at risk of segmentation in Computed Tomography (CT). Many recent studies have focused on several organs such as the lung, heart, esophagus, trachea, liver, aorta, kidney, and prostate. However, among the above organs, the esophagus is one of the most difficult organs to segment because of its small size, ambiguous boundary, and very low contrast in CT images. To address these challenges, we propose a fully automated framework for the esophagus segmentation from CT images. The proposed method is based on the processing of slice images from the original three-dimensional (3D) image so that our method does not require large computational resources. We employ the spatial attention mechanism with the atrous spatial pyramid pooling module to locate the esophagus effectively, which enhances the segmentation performance. To optimize our model, we use group normalization because the computation is independent of batch sizes, and its performance is stable. We also used the simultaneous truth and performance level estimation (STAPLE) algorithm to reach robust results for segmentation. Firstly, our model was trained by k-fold cross-validation. And then, the candidate labels generated by each fold were combined by using the STAPLE algorithm. And as a result, Dice and Hausdorff Distance scores have an improvement when applying this algorithm to our segmentation results. Our method was evaluated on SegTHOR and StructSeg 2019 datasets, and the experiment shows that our method outperforms the state-of-the-art methods in esophagus segmentation. Our approach shows a promising result in esophagus segmentation, which is still challenging in medical analyses.

Prognostic significance of FDG-PET/CT in determining upfront autologous stem cell transplantation for the treatment of peripheral T cell lymphomas.

This observational study aimed to evaluate the prognostic significance of interim and final 18F-fluorodeoxyglucose positron emission tomography-computed tomography (PET/CT) responses to upfront autologous stem cell transplantation (ASCT) in patients with peripheral T cell lymphomas (PTCLs). A total of 118 patients, from two independent institutions, with newly diagnosed PTCLs were enrolled, and 96 of them were evaluated. PET/CT was assessed at diagnosis, and during and after the primary treatment. Clinical outcomes of interim and final PET/CT were compared between transplanted and non-transplanted patients. The responses of PET/CT were assessed based on visual analysis using the Deauville five-point scale (5-PS). Clinicopathological features of transplanted patients (n = 37) were similar to those of non-transplanted patients (n = 59). After a median follow-up of 60.8 months, only final PET/CT response based on 5-PS was the independent prognostic factor of survival outcome (P < 0.001; HR 8.215; 95% C.I. 2.97-22.72) in multivariate analysis. Interim PET/CT response did not have a differential potential for predicting progression-free survival (PFS). In 59 patients, with score 1 or 2 in final PET/CT, the PFS rate was not significantly different between transplanted and non-transplanted patients (P = 0.970). Moreover, among the 37 patients with final PET/CT response score of 3-4, the PFS rate was equally poor in both transplanted and non-transplanted patients (P = 0.178). Final PET/CT assessment, based on 5-PS, was an important prognostic parameter for primary treatment of PTCLs, regardless of upfront ASCT. Interim PET/CT response could not be an indicator to determine the requirement for upfront ASCT.

Association between FDG uptake in the right ventricular myocardium and cancer therapy-induced cardiotoxicity.

BACKGROUND: The aim of this study was to investigate changes in myocardial uptake evaluated by oncologic 18F-fluorodeoxyglucose (FDG) PET/CT scans and to determine the relationship between myocardial FDG uptake and cancer therapy-induced cardiotoxicity in breast cancer patients who underwent anthracycline or trastuzumab. METHODS: We reviewed 121 consecutive patients who underwent oncologic FDG PET/CT and echocardiography at baseline and post-therapy with anthracyclines or trastuzumab for breast cancer. Grade in LV wall, uptake pattern in LV wall, and the presence of RV wall uptake were assessed by visual analysis, and the mean SUV in the LV and RV walls and the change of SUV (ΔSUV) between baseline and post-therapy PET/CT were measured by quantitative analysis. Multiple logistic regression analyses were performed to evaluate the association between PET parameters and cardiotoxicity. RESULTS: Fifteen patients (12%) showed cardiotoxicity after therapy. The cardiotoxic group tended to show more diffuse LV uptake, higher SUV, and ΔSUV of RV wall than the non-cardiotoxic group following therapy with anthracyclines or trastuzumab. Logistic regression analysis showed that the presence of RV wall uptake, SUV of RV wall (> 1.8), and ΔSUV of RV wall (> 0.4) were significantly associated with cardiotoxicity after controlling for age, radiotherapy, and treatment. CONCLUSIONS: The presence of RV wall uptake and the increase of SUV of RV wall on post-therapy PET/CT were associated with cardiotoxicity in breast cancer patients who underwent anthracycline or trastuzumab. Oncologic FDG PET/CT scans can provide information regarding cancer therapy-induced cardiotoxicity as well as tumor response.

The Application of Magnetic Resonance Imaging-Deformed 11C-Methionine-Positron Emission Tomography Images in Stereotactic Radiosurgery.

BACKGROUND: Although 11C-methionine positron emission tomography (MET-PET) images can be fused with magnetic resonance (MR) images using planning software for gamma knife radiosurgery (GKR), the stereotactic information has limited value in patients with recurrent malignant brain tumor due to the difference in imaging protocols between MET-PET and MR images. The aim of this study was to evaluate the clinical application of MR imaging (MRI)-deformed MET-PET images in GKR using a deformable registration tool. METHODS: We examined the enhanced MR stereotactic images, MET-PET and MRI-deformed MET-PET images without stereotactic information for 12 newly developed metastatic brain tumors. MET-PET and MRI-deformed MET-PET images were co-registered with the MR stereotactic images using radiosurgery planning software. Visual analysis was performed to determine whether the MET-PET and MR images matched better after using the deformable registration tool. In addition, the matching volume between MR and MET-PET images was compared before and after applying this tool. The matching volume was calculated as the metabolic tumor volume on the MET-PET images, including the MR-enhanced volume. The matching percentage was calculated as the matching volume divided by the MR-enhanced volume, multiplied by 100. RESULTS: Visual analysis revealed that the MRI-deformed MET-PET images provided the same axial plane as that of the MR images, with the same window level, enabling easy identification of the tumor with the radiosurgery planning software. The mean matching percentage of the MET-PET/MR fusion images was 61.1% (range 24.7-94.7) and that of the MRI-deformed MET-PET/MR fusion images was 63.4% (range 20.8-94.3). No significant difference was found in the matching percentage between the two types of fusion images (p = 0.754). CONCLUSIONS: The MRI-deformed MET-PET images enable utilization of the functional information when planning a treatment in GKR without significant volume change.

The role of interim FDG PET-CT after induction chemotherapy as a predictor of concurrent chemoradiotherapy efficacy and prognosis for head and neck cancer.

PURPOSE: Induction chemotherapy (ICT) with docetaxel, cisplatin, and 5-fluorouracil (TPF) followed by concurrent chemoradiotherapy (CCRT) has the advantages of organ preservation and systemic control in head and neck cancer (HNC). Early prediction of CCRT efficacy may help identify patients who will benefit more from surgery than from CCRT. We investigated the role of interim 18-fluoro-2-deoxy-glucose positron emission tomography computed tomography (FDG PET-CT) after ICT to predict the efficacy of CCRT and clinical outcomes. METHODS: Tumor responses were retrospectively reviewed after CCRT based on the Response Evaluation Criteria in Solid Tumors. FDG PET-CT imaging was performed before and after three cycles of TPF. We examined the associations between the metabolic response (percentage decrease in the maximum standardized uptake value [SUVmax] and total metabolic tumor volume [MTV]) after ICT and complete response (CR) to CCRT, progression-free survival (PFS), and overall survival (OS). RESULTS: We studied 43 HNC patients with a median follow-up of 32.7 months. Lymph node (LN) SUVmax and total MTV decreases from baseline after ICT were greater in patients with a CR to CCRT than in non-CR patients (LN SUVmax, 88.8% vs. 62.5%, respectively; total MTV, 99.7% vs. 89.9%, respectively). Decreases in total MTV ≥ 78% and LN SUVmax ≥73% after ICT predicted CR to CCRT and longer OS and PFS. CONCLUSIONS: Using interim FDG PET-CT to measure SUVmax and total MTV after three cycles of ICT may be a useful technique for identifying HNC patients who will benefit from CCRT and predicting survival outcomes.

Correction to: a risk stratification model for nodal peripheral T-cell lymphomas based on the NCCN-IPI and posttreatment Deauville score.

Unfortunately, the original version of this article contained several errors made during final step of article production. In the results section (fourth sentence) of the Abstract, the incomplete sentence,", 31.4% in high-risk group and 4.7% in treatment failure group.

A risk stratification model for nodal peripheral T-cell lymphomas based on the NCCN-IPI and posttreatment Deauville score.

PURPOSE: The aim of this study was to establish a risk-stratification model integrating posttreatment metabolic response using the Deauville score and the pretreatment National Comprehensive Cancer Network-International Prognostic Index (NCCN-IPI) in nodal PTCLs. METHODS: We retrospectively analysed 326 patients with newly diagnosed nodal PTCLs between January 2005 and June 2016 and both baseline and posttreatment PET/CT data. The final model was validated using an independent prospective cohort of 79 patients. RESULTS: Posttreatment Deauville score (1/2, 3, and 4/5) and the NCCN-IPI (low, low-intermediate, high-intermediate, and high) were independently associated with progression-free survival: for the Deauville score, the hazard ratios (HRs) were 1.00 vs. 2.16 (95% CI 1.47-3.18) vs. 7.86 (5.66-10.92), P < 0.001; and for the NCCN-IPI, the HRs were 1.00 vs. 2.31 (95% CI 1.20-4.41) vs. 4.42 (2.36-8.26) vs. 7.09 (3.57-14.06), P < 0.001. Based on these results, we developed a simplified three-group risk model comprising a low-risk group (low or low-intermediate NCCN-IPI with a posttreatment Deauville score of 1 or 2, or low NCCN-IPI with a Deauville score of 3), a high-risk group (high or high-intermediate NCCN-IPI with a Deauville score of 1/2 or 3, or low-intermediate NCCN-IPI with a Deauville score of 3), and a treatment failure group (Deauville score 4 or 5). This model was significantly associated with progression-free survival (5-year, 70.3%, 31.4%, and 4.7%; P < 0.001) and overall survival (5-year, 82.1%, 45.5%, and 14.7%; P < 0.001). Similar associations were also observed in the independent validation cohort. CONCLUSION: The risk-stratification model integrating posttreatment Deauville score and pretreatment NCCN-IPI is a powerful tool for predicting treatment failure in patients with nodal PTCLs.

Prediction of coronary artery calcium progression by FDG uptake of large arteries in asymptomatic individuals.

PURPOSE: The purpose of this study is to evaluate whether fluorodeoxyglucose (FDG) uptake of the large arteries can predict coronary artery calcium (CAC) progression in asymptomatic individuals. METHODS: Ninety-six asymptomatic individuals who underwent FDG positron emission tomography (PET) and CAC scoring on the same day for health screening and follow-up CAC scoring ≥1 year after baseline studies (mean 4.3 years) were included. Vascular FDG uptake was measured and corrected for blood pool activity to obtain peak and average target-to-blood pool ratios (TBRpeak and TBRavg, respectively) for the carotid arteries, and ascending and abdominal aorta. CAC scores at baseline and follow-up of each individual were measured and absolute CAC change (ΔCAC), annual CAC change (ΔCAC/year), and annual CAC change rate (ΔCAC%/year) were calculated. CAC progression was defined as ΔCAC >0 for individuals with negative baseline CAC; ΔCAC/year ≥10 for those with baseline CAC of 0

Clinical values of left ventricular mechanical dyssynchrony assessment by gated myocardial perfusion SPECT in patients with acute myocardial infarction and multivessel disease.

PURPOSE: The aim of this study was to evaluate the prognostic value of additional evaluation of left ventricular mechanical dyssynchrony (LVMD) by gated myocardial perfusion single-photon emission computed tomography (GMPS) in patients with acute myocardial infarction (MI) and multivessel disease. METHODS: One hundred and nine acute MI patients with >50 % stenosis in at least one non-culprit artery who underwent GMPS within 2 weeks were enrolled. All patients underwent successful revascularization of the culprit arteries. Those with previous MI, atrial fibrillation, or frequent ventricular premature complexes, cardiac devices, significant patient motion, or procedure-related events were excluded. Phase standard deviation (PSD) and phase histogram bandwidth (PBW) were measured for assessment of LVMD. Patients were followed up for a median of 26 months after index MI, for composite major adverse cardiac events (MACE), which consisted with all-cause death, unplanned hospitalization due to heart failure and severe ventricular arrhythmias (sustained ventricular tachycardia or ventricular fibrillation). Independent predictors of MACE were evaluated. RESULTS: MACE occurred in 22 patients (20 %). Stress PSD (53.3 ± 17.3° vs. 35.3 ± 18.9°; p <0.001), stress PBW (147.6 ± 54.6° vs. 96.8 ± 59.2°; p = 0.001) and resting PBW (126.8 ± 37.5° vs. 96.6 ± 48.9°; p = 0.001) were significantly higher in patients with MACE compared to those without. Multivariate analysis revealed that stress PSD ≥45.5° and stress PBW ≥126.0° were predictive of MACE, as well as suboptimal non-culprit artery revascularization (SNR) and renin-angiotensin system (RAS) blockade medication. Higher stress PSD and stress PBW were associated with poorer prognosis both in patients with and without SNR, and those with RAS blockade medication, but not in those without RAS blockade medication. CONCLUSIONS: LVMD measured by GMPS showed added prognostic value in acute MI with multivessel disease. GMPS could serve as a comprehensive evaluation imaging tool in patients with acute MI and multivessel disease.

Clinical implication of F-18 FDG PET/CT in patients with secondary hemophagocytic lymphohistiocytosis.

The contribution that F-18 fluoro-2-deoxyglucose positron emission tomography/computed tomography (F-18 FDG) PET/CT makes to the diagnosis of malignancy in patients with hemophagocytic lymphohistiocytosis (HLH) is still uncertain. The aim of this study was to evaluate the diagnostic performance of F-18 FDG PET/CT for the detection of underlying malignancy, to investigate the correlation between PET and laboratory parameters, and to identify prognosis-related factors in patients with secondary HLH. We enrolled 14 patients who were diagnosed with HLH and referred for F-18 FDG PET/CT to exclude malignancy. The diagnostic performance of F-18 FDG PET/CT for malignancy detection was assessed. The correlations between PET and laboratory parameters were determined. The prognostic significance of the following factors was evaluated: PET and laboratory parameters, age in years, presence of underlying malignancy, and fever and splenomegaly. Six of the 14 patients had malignancies (four with lymphoma, one with multiple myeloma, and one with colonic malignancy). Sensitivity, specificity, and diagnostic accuracy of F-18 FDG PET/CT for malignancy detection were 83, 62.5, and 71.4 %, respectively. F-18 FDG uptake in the bone marrow and spleen was positively correlated with neutrophil count and C-reactive protein. All of the PET parameters, but none of the clinical or laboratory parameters, were significantly associated with patient outcome, as determined by univariate analysis. Given the small sample size, F-18 FDG PET/CT was useful for detecting underlying malignancy, and PET parameters correlated with laboratory parameters that reflected inflammatory status. F-18 FDG PET/CT might provide prognostic information for the management of patients with secondary HLH.

A comparison between centralized and asynchronous federated learning approaches for survival outcome prediction using clinical and PET data from non-small cell lung cancer patients.

BACKGROUND AND OBJECTIVE: Survival analysis plays an essential role in the medical field for optimal treatment decision-making. Recently, survival analysis based on the deep learning (DL) approach has been proposed and is demonstrating promising results. However, developing an ideal prediction model requires integrating large datasets across multiple institutions, which poses challenges concerning medical data privacy. METHODS: In this paper, we propose FedSurv, an asynchronous federated learning (FL) framework designed to predict survival time using clinical information and positron emission tomography (PET)-based features. This study used two datasets: a public radiogenic dataset of non-small cell lung cancer (NSCLC) from the Cancer Imaging Archive (RNSCLC), and an in-house dataset from the Chonnam National University Hwasun Hospital (CNUHH) in South Korea, consisting of clinical risk factors and F-18 fluorodeoxyglucose (FDG) PET images in NSCLC patients. Initially, each dataset was divided into multiple clients according to histological attributes, and each client was trained using the proposed DL model to predict individual survival time. The FL framework collected weights and parameters from the clients, which were then incorporated into the global model. Finally, the global model aggregated all weights and parameters and redistributed the updated model weights to each client. We evaluated different frameworks including single-client-based approach, centralized learning and FL. RESULTS: We evaluated our method on two independent datasets. First, on the RNSCLC dataset, the mean absolute error (MAE) was 490.80±22.95 d and the C-Index was 0.69±0.01. Second, on the CNUHH dataset, the MAE was 494.25±40.16 d and the C-Index was 0.71±0.01. The FL approach achieved centralized method performance in PET-based survival time prediction and outperformed single-client-based approaches. CONCLUSIONS: Our results demonstrated the feasibility and effectiveness of employing FL for individual survival prediction in NSCLC patients, using clinical information and PET-based features.

Prediction of immunochemotherapy response for diffuse large B-cell lymphoma using artificial intelligence digital pathology.

Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous and prevalent subtype of aggressive non-Hodgkin lymphoma that poses diagnostic and prognostic challenges, particularly in predicting drug responsiveness. In this study, we used digital pathology and deep learning to predict responses to immunochemotherapy in patients with DLBCL. We retrospectively collected 251 slide images from 216 DLBCL patients treated with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP), with their immunochemotherapy response labels. The digital pathology images were processed using contrastive learning for feature extraction. A multi-modal prediction model was developed by integrating clinical data and pathology image features. Knowledge distillation was employed to mitigate overfitting on gigapixel histopathology images to create a model that predicts responses based solely on pathology images. Based on the importance derived from the attention mechanism of the model, we extracted histological features that were considered key textures associated with drug responsiveness. The multi-modal prediction model achieved an impressive area under the ROC curve of 0.856, demonstrating significant associations with clinical variables such as Ann Arbor stage, International Prognostic Index, and bulky disease. Survival analyses indicated their effectiveness in predicting relapse-free survival. External validation using TCGA datasets supported the model's ability to predict survival differences. Additionally, pathology-based predictions show promise as independent prognostic indicators. Histopathological analysis identified centroblastic and immunoblastic features to be associated with treatment response, aligning with previous morphological classifications and highlighting the objectivity and reproducibility of artificial intelligence-based diagnosis. This study introduces a novel approach that combines digital pathology and clinical data to predict the response to immunochemotherapy in patients with DLBCL. This model shows great promise as a diagnostic and prognostic tool for clinical management of DLBCL. Further research and genomic data integration hold the potential to enhance its impact on clinical practice, ultimately improving patient outcomes.

Ensemble Deep Learning Model to Predict Lymphovascular Invasion in Gastric Cancer.

Lymphovascular invasion (LVI) is one of the most important prognostic factors in gastric cancer as it indicates a higher likelihood of lymph node metastasis and poorer overall outcome for the patient. Despite its importance, the detection of LVI(+) in histopathology specimens of gastric cancer can be a challenging task for pathologists as invasion can be subtle and difficult to discern. Herein, we propose a deep learning-based LVI(+) detection method using H&E-stained whole-slide images. The ConViT model showed the best performance in terms of both AUROC and AURPC among the classification models (AUROC: 0.9796; AUPRC: 0.9648). The AUROC and AUPRC of YOLOX computed based on the augmented patch-level confidence score were slightly lower (AUROC: -0.0094; AUPRC: -0.0225) than those of the ConViT classification model. With weighted averaging of the patch-level confidence scores, the ensemble model exhibited the best AUROC, AUPRC, and F1 scores of 0.9880, 0.9769, and 0.9280, respectively. The proposed model is expected to contribute to precision medicine by potentially saving examination-related time and labor and reducing disagreements among pathologists.

Prognostic Significance Of Sequential 18f-fdg Pet/Ct During Frontline Treatment Of Peripheral T Cell Lymphomas.

BACKGROUND/AIMS: The prognostic significance of 18F-fluorodeoxyglucose (FDG)-positron emission tomography-computed tomography (PET/CT) in peripheral T-cell lymphomas (PTCLs) are controversial. We explored the prognostic impact of sequential 18F-FDG PET/CT during frontline chemotherapy of patients with PTCLs. METHODS: In total, 143 patients with newly diagnosed PTCLs were included. Sequential 18F-FDG PET/CTs were performed at the time of diagnosis, during chemotherapy, and at the end of chemotherapy. The baseline total metabolic tumor volume (TMTV) was calculated using the the standard uptake value with a threshold method of 2.5. RESULTS: A baseline TMTV of 457.0 cm3 was used to categorize patients into high and low TMTV groups. Patients with a requirehigh TMTV had shorter progression-free survival (PFS) and overall survival (OS) than those with a low TMTV (PFS, 9.8 vs. 26.5 mo, p = 0.043; OS, 18.9 vs. 71.2 mo, p = 0.004). The interim 18F-FDG PET/CT response score was recorded as 1, 2-3, and 4-5 according to the Deauville criteria. The PFS and OS showed significant differences according to the interim 18F-FDG PET/CT response score (PFS, 120.7 vs. 34.1 vs. 5.1 mo, p < 0.001; OS, not reached vs. 61.1 mo vs. 12.1 mo, p < 0.001). CONCLUSION: The interim PET/CT response based on visual assessment predicts disease progression and survival outcome in PTCLs. A high baseline TMTV is associated with a poor response to anthracycline-based chemotherapy in PTCLs. However, TMTV was not an independent predictor for PFS in the multivariate analysis.

Survival Prediction Using Transformer-Based Categorical Feature Representation in the Treatment of Diffuse Large B-Cell Lymphoma.

Diffuse large B-cell lymphoma (DLBCL) is a common and aggressive subtype of lymphoma, and accurate survival prediction is crucial for treatment decisions. This study aims to develop a robust survival prediction strategy to integrate various risk factors effectively, including clinical risk factors and Deauville scores in positron-emission tomography/computed tomography at different treatment stages using a deep-learning-based approach. We conduct a multi-institutional study on 604 DLBCL patients' clinical data and validate the model on 220 patients from an independent institution. We propose a survival prediction model using transformer architecture and a categorical-feature-embedding technique that can handle high-dimensional and categorical data. Comparison with deep-learning survival models such as DeepSurv, CoxTime, and CoxCC based on the concordance index (C-index) and the mean absolute error (MAE) demonstrates that the categorical features obtained using transformers improved the MAE and the C-index. The proposed model outperforms the best-performing existing method by approximately 185 days in terms of the MAE for survival time estimation on the testing set. Using the Deauville score obtained during treatment resulted in a 0.02 improvement in the C-index and a 53.71-day improvement in the MAE, highlighting its prognostic importance. Our deep-learning model could improve survival prediction accuracy and treatment personalization for DLBCL patients.

Molecular imaging approaches to facilitate bacteria-mediated cancer therapy.

Bacteria-mediated cancer therapy is a potential therapeutic strategy for cancer that has unique properties, including broad tumor-targeting ability, various administration routes, the flexibility of delivery, and facilitating the host's immune responses. The molecular imaging of bacteria-mediated cancer therapy allows the therapeutically injected bacteria to be visualized and confirms the accurate delivery of the therapeutic bacteria to the target lesion. Several hurdles make bacteria-specific imaging challenging, including the need to discriminate therapeutic bacterial infection from inflammation or other pathologic lesions. To realize the full potential of bacteria-specific imaging, it is necessary to develop bacteria-specific targets that can be associated with an imaging assay. This review describes the current status of bacterial imaging techniques together with the advantages and disadvantages of several imaging modalities. Also, we describe potential targets for bacterial-specific imaging and related applications.