Computed tomography-based radiomics combined with machine learning allows differentiation between primary intestinal lymphoma and Crohn's disease.

BACKGROUND: Due to similar clinical manifestations and imaging signs, differential diagnosis of primary intestinal lymphoma (PIL) and Crohn's disease (CD) is a challenge in clinical practice. AIM: To investigate the ability of radiomics combined with machine learning methods to differentiate PIL from CD. METHODS: We collected contrast-enhanced computed tomography (CECT) and clinical data from 120 patients form center 1. A total of 944 features were extracted single-phase images of CECT scans. Using the last absolute shrinkage and selection operator model, the best predictive radiographic features and clinical indications were screened. Data from 54 patients were collected at center 2 as an external validation set to verify the robustness of the model. The area under the receiver operating characteristic curve, accuracy, sensitivity and specificity were used for evaluation. RESULTS: A total of five machine learning models were built to distinguish PIL from CD. Based on the results from the test group, most models performed well with a large area under the curve (AUC) (> 0.850) and high accuracy (> 0.900). The combined clinical and radiomics model (AUC = 1.000, accuracy = 1.000) was the best model among all models. CONCLUSION: Based on machine learning, a model combining clinical data with radiologic features was constructed that can effectively differentiate PIL from CD.

[Analysis on the ultrasonic characteristics of tonsillar lymphoma].

Objective: To analyze the ultrasonic features of tonsillar lymphoma to improve the diagnostic accuracy. Methods: The clinical, pathological and ultrasonic data of nine patients with tonsillar lymphoma confirmed by pathology at Tianjin Medical University Cancer Institute and Hospital during June 2015 and June 2022 were analyzed retrospectively, and the characteristics of their ultrasonic images were summarized. Results: All 9 cases of tonsil lymphoma were unilateral tonsil disease, including 4 cases on the left side and 5 cases on the right side. The average maximum diameter of tonsil lymphoma in 9 cases was 4.32 cm. There were 3 cases with simultaneous involvement of tonsil and cervical lymph nodes, all of which were ipsilateral lymph nodes. Gray scale ultrasound showed that the lesions were hypoechoic, with clear boundaries in 7 cases and unclear boundaries in 2 cases. The shape was full and irregular in 5 cases and oval in 4 cases. The echo was uniform in 7 cases and uneven in 2 cases. Color Doppler ultrasonography showed abundant internal blood flow signal in 1 case, a little dotted linear internal blood flow signal in 5 cases, and no obvious internal blood flow signal in 3 cases. Conclusions: The ultrasonic features of tonsillar lymphoma include hypoechoic area, clear boundary, full shape, irregular and uniform internal echo, no or low linear signal of internal blood flow. Ultrasonography is of great value in the diagnosis of this disease and can help clinical decision-making.

Preliminary study on the ability of the machine learning models based on 18F-FDG PET/CT to differentiate between mass-forming pancreatic lymphoma and pancreatic carcinoma.

PURPOSE: The objective of this study was to preliminarily assess the ability of metabolic parameters and radiomics derived from 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) to distinguish mass-forming pancreatic lymphoma from pancreatic carcinoma using machine learning. METHODS: A total of 88 lesions from 86 patients diagnosed as mass-forming pancreatic lymphoma or pancreatic carcinoma were included and randomly divided into a training set and a validation set at a 4-to-1 ratio. The segmentation of regions of interest was performed using ITK-SNAP software, PET metabolic parameters and radiomics features were extracted using 3Dslicer and PYTHON. Following the selection of optimal metabolic parameters and radiomics features, Logistic regression (LR), support vector machine (SVM), and random forest (RF) models were constructed for PET metabolic parameters, CT radiomics, PET radiomics, and PET/CT radiomics. Model performance was assessed in terms of area under the curve (AUC), accuracy, sensitivity, and specificity in both the training and validation sets. RESULTS: Strong discriminative ability observed in all models, with AUC values ranging from 0.727 to 0.978. The highest performance exhibited by the combined PET and CT radiomics features. AUC values for PET/CT radiomics models in the training set were LR 0.994, SVM 0.994, RF 0.989. In the validation set, AUC values were LR 0.909, SVM 0.883, RF 0.844. CONCLUSION: Machine learning models utilizing the metabolic parameters and radiomics of 18F-FDG PET/CT show promise in distinguishing between pancreatic carcinoma and mass-forming pancreatic lymphoma. Further validation on a larger cohort is necessary before practical implementation in clinical settings.

Synthetic MRI and diffusion-weighted imaging for differentiating nasopharyngeal lymphoma from nasopharyngeal carcinoma: combination with morphological features.

OBJECTIVES: To investigate the feasibility of synthetic MRI (syMRI), diffusion-weighted imaging (DWI), and their combination with morphological features for differentiating nasopharyngeal lymphoma (NPL) from nasopharyngeal carcinoma (NPC). METHODS: Sixty-nine patients with nasopharyngeal tumours (NPL, n = 22; NPC, n = 47) who underwent syMRI and DWI were retrospectively enrolled between October 2020 and May 2022. syMRI and DWI quantitative parameters (T1, T2, PD, ADC) and morphological features were obtained. Diagnostic performance was assessed by independent sample t-test, chi-square test, logistic regression analysis, receiver operating characteristic curve (ROC), and DeLong test. RESULTS: NPL has significantly lower T2, PD, and ADC values compared to NPC (all P < .05), whereas no significant difference was found in T1 value between these two entities (P > .05). The morphological features of tumour type, skull-base involvement, Waldeyer ring involvement, and lymph nodes involvement region were significantly different between NPL and NPC (all P < .05). The syMRI (T2 + PD) model has better diagnostic efficacy, with AUC, sensitivity, specificity, and accuracy of 0.875, 77.27%, 89.36%, and 85.51%. Compared with syMRI model, syMRI + Morph (PD + Waldeyer ring involvement + lymph nodes involvement region), syMRI + DWI (T2 + PD + ADC), and syMRI + DWI + Morph (PD + ADC + skull-base involvement + Waldeyer ring involvement) models can further improve the diagnostic efficiency (all P < .05). Furthermore, syMRI + DWI + Morph model has excellent diagnostic performance, with AUC, sensitivity, specificity, and accuracy of 0.986, 95.47%, 97.87%, and 97.10%, respectively. CONCLUSION: syMRI and DWI quantitative parameters were helpful in discriminating NPL from NPC. syMRI + DWI + Morph model has the excellent diagnostic efficiency in differentiating these two entities. ADVANCES IN KNOWLEDGE: syMRI + DWI + morphological feature method can differentiate NPL from NPC with excellent diagnostic performance.

Fluorine-18 fluorodeoxyglucose uptake change in liver, mediastinal blood pool, and lymphoid cell-rich organs during programmed cell death-1 immunotherapy in lymphoma.

PURPOSE: The aim of this study was to evaluate metabolism change in reference organs (liver and mediastinum) and lymphoid cell-rich organs (spleen and bone marrow) during programmed cell death-1 immunotherapy in relapsed or refractory lymphoma patients. METHODS: A total of 66 patients with baseline and serial monitoring fluorodeoxyglucose (FDG) PET/computed tomography scans were retrospectively enrolled. Mean standardized uptake value (SUV) and maximum SUV of evaluated organs were obtained by two reviewers, and their association with tumor burden and clinical response were evaluated. Immune-related adverse events detected by FDG PET/computed tomography were also recorded. RESULTS: The SUV values of reference organs and lymphoid cell-rich organs did not change significantly during the immunotherapy process. The intersubject variability of these values ranged from 13.0 to 28.5%. Meanwhile, metabolism of reference organs was affected by neither the tumor burden nor clinical response. SUV change of lymphoid cell-rich organs was associated with clinical response to immunotherapy. Responders showed decreased metabolism, while nonresponders showed a reverse trend (spleen SUV max : -0.30 ±â€…0.47 vs. 0.18 ±â€…0.39, P  = 0.001, spleen SUV mean : -0.24 ±â€…0.39 vs. 0.14 ±â€…0.31, P  = 0.001; and bone marrow SUV max : -0.14 ±â€…0.37 vs. 0.07 ±â€…0.46, P  = 0.042, respectively). The influence of immune-related adverse events on the SUV change in evaluated organs was not significant. CONCLUSION: During programmed cell death-1 immunotherapy, metabolism change of reference organs is influenced neither by tumor burden nor by clinical response, while FDG uptake change of lymphoid cell-rich organs is significantly associated with clinical response.

Lymphoma presenting as preauricular tumor in unilateral parotid gland agenesis: a case report and review of literature.

BACKGROUND: Parotid gland agenesis is a rare, congenital, usually asymptomatic disorder. Until now, only 24 cases with unilateral, incidentally found, parotid gland agenesis have been described. Here, we present the first reported case of an ipsilateral preauricular neoplasm in a patient with unilateral parotid gland agenesis. During surgery, the position of the greater auricular- and facial nerves was documented. Furthermore, we performed the first sialendoscopy for this rare disorder to assess the number of duct branches, which might be indicative of the abundance of parotid tissue. Moreover, we looked for sialendoscopic characteristic features that could aid in identifying these patients in the ambulatory setting. CASE PRESENTATION: A 50-year-old Greek man presented with a painless, slowly enlarging mass in the right parotid space. Magnetic resonance imaging revealed a complete absence of the right parotid gland without accessory parotid tissue. The right parotid gland was replaced by fatty tissue and the radiologist suggested a benign parotid tumor. Fine needle aspiration was indicative of a reactive lymph node. Sialendoscopy revealed only two branches within the right parotid duct. Surgical resection was performed through a conventional lateral parotidectomy. This revealed typical anatomic position of the greater auricular- and facial nerves despite the parotid tissue agenesis. Histopathology revealed a small lymphocytic lymphoma. CONCLUSIONS: Surgeons should feel confident to resect tumors of the parotid space in patients with parotid gland agenesis. Reduced branching observed during sialendoscopy might indicate parotid gland agenesis. Physicians should be even more cautious than usual with the watch and wait strategy in patients with tumors of parotid gland agenesis, since the probability of a tumor being a benign salivary gland tumor might be lower than usual.

Diagnosis of cervical lymphoma using a YOLO-v7-based model with transfer learning.

To investigate the ability of an auxiliary diagnostic model based on the YOLO-v7-based model in the classification of cervical lymphadenopathy images and compare its performance against qualitative visual evaluation by experienced radiologists. Three types of lymph nodes were sampled randomly but not uniformly. The dataset was randomly divided into for training, validation, and testing. The model was constructed with PyTorch. It was trained and weighting parameters were tuned on the validation set. Diagnostic performance was compared with that of the radiologists on the testing set. The mAP of the model was 96.4% at the 50% intersection-over-union threshold. The accuracy values of it were 0.962 for benign lymph nodes, 0.982 for lymphomas, and 0.960 for metastatic lymph nodes. The precision values of it were 0.928 for benign lymph nodes, 0.975 for lymphomas, and 0.927 for metastatic lymph nodes. The accuracy values of radiologists were 0.659 for benign lymph nodes, 0.836 for lymphomas, and 0.580 for metastatic lymph nodes. The precision values of radiologists were 0.478 for benign lymph nodes, 0.329 for lymphomas, and 0.596 for metastatic lymph nodes. The model effectively classifies lymphadenopathies from ultrasound images and outperforms qualitative visual evaluation by experienced radiologists in differential diagnosis.

Generating lymphoma ultrasound image description with transformer model.

Lymphoma, the most prevalent hematologic tumor originating from the lymphatic hematopoietic system, can be accurately diagnosed using high-resolution ultrasound. Microscopic ultrasound performance enables clinicians to identify suspected tumors and subsequently obtain a definitive pathological diagnosis through puncture biopsy. However, the complex and diverse ultrasonographic manifestations of lymphoma pose challenges for accurate characterization by sonographers. To address these issues, this study proposes a Transformer-based model for generating descriptive ultrasound images of lymphoma, aiming to provide auxiliary guidance for ultrasound doctors during screening procedures. Specifically, deep stable learning is integrated into the model to eliminate feature dependencies by training sample weights. Additionally, a memory module is incorporated into the model decoder to enhance semantic information modeling in descriptions and utilize learned semantic tree branch structures for more detailed image depiction. Experimental results on an ultrasonic diagnosis dataset from Shanghai Ruijin Hospital demonstrate that our proposed model outperforms relevant methods in terms of prediction performance.

Update on FDG-PET in pediatric lymphoma.

Lymphoma represent the third most common malignant disease in childhood and adolescence. They are divided into pediatric Hodgkin lymphoma (P-HL) and pediatric non-Hodgkin lymphoma (P-NHL). In P-HL, excellent cure rates are achieved through combined modality treatment using chemotherapy and radiotherapy. For more than 20 years, FDG-PET has been an integral part of the treatment and guides its intensity through improved staging and precise assessment of chemotherapy response. In P-NHL, good cure rates are achieved with chemotherapy alone. At present FDG-PET plays only a subordinate role in the treatment setting. Its potential to contribute to treatment management is far from being fully utilised. In this article, the current status of FDG-PET in pediatric lymphoma is presented in detail. The core elements are the sections on staging and response assessment. In addition, challenges and pitfalls are discussed and future developments are outlined.

Distinguishing Kikuchi-Fujimoto disease from lymphoma in patients by clinical and PET/CT features.

To develop a scheme for distinguishing Kikuchi-Fujimoto disease (KFD) from lymphoma in patients presenting enlarged lymph nodes (LNs) predominantly on the upper side of the diaphragm. From November 2015 to August 2023, 32 KFD patients and 38 lymphoma patients were pathologically confirmed and enrolled in this retrospectively study. Clinical and 18F-fluorodeoxyglucose positron emission tomography (PET)/computed tomography (CT) features were collected. When comparing those PET/CT parameters, we set 5 models with different research objects: (1) all affected LNs; (2) the 5 largest affected LNs in terms of maximum diameter; (3) the 5 largest affected LNs in terms of maximum standard uptake values (SUVmax); (4) the largest affected LNs in terms of maximum diameter; (5) the largest affected LNs in terms of SUVmax. Compared to lymphoma patients, KFD patients were younger; and with higher incidence of fever, arthralgia, abnormal serum white blood cell, lactate dehydrogenase (LDH) and splenomegaly; lower incidence of affected LNs perinodal infiltration, necrosis and conglomeration; more affected LNs in Head and Neck nodes (particularly in level II) and Axillary in KFD (P ˂ .05). PET/CT parameters presented as various difference in each model. Finally, 11 clinical and PET/CT features (age ≤ 34, with fever, arthralgia, abnormal white blood cell, abnormal LDH, and without node necrosis and node conglomeration have a score of 2 each; splenomegaly, perinodal infiltration, median maximum diameter ≤ 20.5 and median SUVmax ≤ 7.1 of affected LNs in model 2 have score of 1 each) were selected as scheme items for distinguishing KFD from lymphoma. Individuals who have a total score > 8, meet the criteria for KFD. Sensitivity and specificity were high: 86.8% (95% CI: 71.9%, 95.5%) and 96.9% (95% CI: 83.7%, 99.5%), AUC = 0.975 (95% CI: 90.5%, 99.6%), respectively. It can effectively distinguish KFD from lymphoma by clinical and PET/CT parameters.

Sociodemographic factors and Child Opportunity Index disparities associated with missed care opportunities in pediatric patients with lymphoma and leukemia referred for FDG-PET/CT.

BACKGROUND: Little data exists on the association of missed care opportunities (MCOs) in children referred for nuclear medicine/nuclear oncology imaging examinations and socioeconomic disparities. OBJECTIVE: To determine the prevalence of MCOs in children with lymphoma/leukemia scheduled for fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT) and the impact of sociodemographic factors and Child Opportunity Index (COI). MATERIALS AND METHODS: Retrospective analysis of MCOs in children with lymphoma/leukemia scheduled for FDG-PET/CT (2012 to 2022) was performed. In univariate analysis, patient, neighborhood, and appointment data were assessed across MCOs and completed appointments. Logistic regression evaluated independent effects of patient-, neighborhood-, and appointment-level factors with MCOs. Two-sided P-value < .05 was considered statistically significant. RESULTS: In 643 FDG-PET/CT appointments (n = 293 patients; median age 15 years (IQR 11.0-17.0 years); 37.9% female), there were 20 MCOs (3.1%) involving 16 patients. Only 8.2% appointments involved Black/African American non-Hispanic/Latino patients, yet they made up a quarter of total MCOs. Patients living in neighborhoods with very low or low COI experienced significantly higher MCOs versus zip codes with very high COI (6.9% vs. 0.8%; P = 0.02). Logistic regression revealed significantly increased likelihood of MCOs for patients aged 18 to 21 [odds ratio (OR) 4.50; 95% CI 1.53-13.27; P = 0.007], Black/African American non-Hispanic/Latino (OR 3.20; 95% CI 1.08-9.49; P = 0.04), zip codes with very low or low COI (OR 9.60; 95% CI 1.24-74.30; P = 0.03), and unknown insurance status. CONCLUSION: Children with lymphoma/leukemia, living in zip codes with very low or low COI, and who identified as Black/African American non-Hispanic/Latino experienced more MCOs. Our study supports the need to address intersecting sociodemographic, neighborhood, and health system factors that will improve equitable access to necessary healthcare imaging for children.

How does deep learning/machine learning perform in comparison to radiologists in distinguishing glioblastomas (or grade IV astrocytomas) from primary CNS lymphomas?: a meta-analysis and systematic review.

BACKGROUND: Several studies have been published comparing deep learning (DL)/machine learning (ML) to radiologists in differentiating PCNSLs from GBMs with equivocal results. We aimed to perform this meta-analysis to evaluate the diagnostic accuracy of ML/DL versus radiologists in classifying PCNSL versus GBM using MRI. METHODOLOGY: The study was performed in accordance with PRISMA guidelines. Data was extracted and interpreted by two researchers with 12 and 23 years' experience, respectively, and QUADAS-2 tool was used for quality and risk-bias assessment. We constructed contingency tables to derive sensitivity, specificity accuracy, summary receiver operating characteristic (SROC) curve, and the area under the curve (AUC). RESULTS: Our search identified 11 studies, of which 8 satisfied our inclusion criteria and restricted the analysis in each study to reporting the model showing highest accuracy, with a total sample size of 1159 patients. The random effects model showed a pooled sensitivity of 0.89 [95% CI:0.84-0.92] for ML and 0.82 [95% CI:0.76-0.87] for radiologists. Pooled specificity was 0.88 [95% CI: 0.84-0.91] for ML and 0.90 [95% CI: 0.81-0.95] for radiologists. Pooled accuracy was 0.88 [95% CI: 0.86-0.90] for ML and 0.86 [95% CI: 0.78-0.91] for radiologists. Pooled AUC of ML was 0.94 [95% CI:0.92-0.96]and for radiologists, it was 0.90 [95% CI: 0.84-0.93]. CONCLUSIONS: MRI-based ML/DL techniques can complement radiologists to improve the accuracy of classifying GBMs from PCNSL, possibly reduce the need for a biopsy, and avoid any unwanted neurosurgical resection of a PCNSL.

Interim FDG-PET improves treatment failure prediction in primary central nervous system lymphoma: An LOC network prospective multicentric study.

BACKGROUND: The purpose of our study was to assess the predictive and prognostic role of 2-18F-fluoro-2-deoxy-d-glucose (FDG) positron emission tomography (PET)/MRI during high-dose methotrexate-based chemotherapy (HD-MBC) in de novo primary central nervous system lymphoma (PCNSL) patients aged 60 and above. METHODS: This prospective multicentric ancillary study included 65 immunocompetent patients who received induction HD-MBC as part of the BLOCAGE01 phase III trial. FDG-PET/MRI were acquired at baseline, post 2 cycles (PET/MRI2), and posttreatment (PET/MRI3). FDG-PET response was dichotomized with "positive" indicating persistent tumor uptake higher than the contralateral mirroring brain region. Performances of FDG-PET and International PCNSL Collaborative Group criteria in predicting induction response, progression-free survival (PFS), and overall survival (OS) were compared. RESULTS: Of the 48 PET2 scans performed, 9 were positive and aligned with a partial response (PR) on MRI2. Among these, 8 (89%) progressed by the end of the induction phase. In contrast, 35/39 (90%) of PET2-negative patients achieved complete response (CR). Among the 18 discordant responses at interim (PETCR/MRIPR), 83% ultimately achieved CR. Eighty-seven percent of the PET2-negative patients were disease free at 6 months versus 11% of the PET2-positive patients (P < .001). The MRI2 response did not significantly differentiate patients based on their PFS, regardless of whether they were in CR or PR. Both PET2 and MRI2 independently predicted OS in multivariate analysis, with PET2 showing a stronger association. CONCLUSIONS: Our study highlights the potential of interim FDG-PET for early management of PCNSL patients. Response-driven treatment based on PET2 may guide future clinical trials. TRIAL: LOCALYZE, NCT03582254, ancillary of phase III clinical trial BLOCAGE01, NCT02313389 (Registered July 10, 2018-retrospectively registered) https://clinicaltrials.gov/ct2/show/NCT03582254?term=LOCALYZE&draw=2&rank=1.

Quarter-Century PET/CT Transformation of Oncology: Lymphoma.

The clinical landscape of lymphomas has changed dramatically over the last 2 decades, including significant progress made in the understanding and utilization of imaging modalities and the available treatment options for both indolent and aggressive lymphomas. Since the introduction of hybrid PET/CT scanners in 2001, the indications of 18F-fluorodeoxyglucose (FDG) PET/CT in the management of lymphomas have grown rapidly. In today's clinical practice, FDG PET/CT is used in successful management of the vast majority patients with lymphomas.