Radiomics Features of the Spleen as Surrogates for CT-Based Lymphoma Diagnosis and Subtype Differentiation

Enke, Johanna S.; Moltz, Jan Hendrik; D’Anastasi, Melvin; Kunz, Wolfgang G.; Schmidt, Christian; Maurus, Stefan; Mühlberg, Alexander; Katzmann, Alexander; Sühling, Michael; Hahn, Horst K.; Nörenberg, Dominik; Huber, Thomas S.

doi:10.3390/cancers14030713

Cited by 16 publications

(11 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conversely, the degree of splenomegaly was reported as an inappropriate index in previous studies. [1,27] The splenomegaly had been defined as >13 cm in length in the long axis. [1,4] The feasibility of 13 cm in length for criteria of splenic involvement was investigated by Berzaczy et al, showing an accuracy of 74.1% and a specificity of 47%.…”

Section: Discussionmentioning

confidence: 99%

Optimal index for detecting splenic involvement on 18F-fluorodeoxyglucose positron emission tomography/computed tomography imaging in diffuse large B-cell lymphoma

Kozuki,

Sugimoto,

Goto

et al. 2024

Medicine

View full text Add to dashboard Cite

Accurate clinical staging is important in diffuse large B-cell lymphoma (DLBCL) to adapt to optimal therapy. Splenic involvement of DLBCL has been recently more detectable with the advancement of a diagnostic scan by 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET)/computed tomography (CT). Our clinical question is whether splenic involvement was adequately diagnosed by FDG-PET/CT imaging. This retrospective study aimed to determine the optimal index for evaluating splenic involvement in patients with DLBCL. Patients with newly diagnosed DLBCL who were examined with FDG-PET/CT at diagnosis and the end of induction chemotherapy (EOI) was enrolled. The splenic involvement with the splenic FDG uptake value higher than that of the liver at diagnosis or with the decrease of splenic uptake at EOI by visual evaluation was evaluated as positive. The calculative evaluation of splenic involvement, based on the data of standardized uptake value (SUV) of the spleen, used maximum SUV (SUVmax), mean SUV (SUVmean), spleen total lesion glycolysis (spleen TLG), and spleen length. A change in each index following induction chemotherapy was expressed as an index. Receiver operating characteristic analysis was used to set the cutoff value for each index. This study included 52 patients. Spleen TLG (0.904) showed the best accuracy, followed by SUVmax (0.885) and SUVmean (0.885), among the 5 indexes for splenic involvement at diagnosis. Splenic involvement was predicted with a higher accuracy level (0.923) when selecting the cases with values higher than the cutoff level on both spleen TLG and SUVmax. The decision at EOI was more suitable by selecting both positive cases of ∆ TLG and ∆ SUVmax. Obtaining both the positive spleen TLG and SUVmax is recommended at diagnosis to predict splenic involvement. The assessment by ∆ spleen TLG and ∆ SUVmax seems to be optimal.

show abstract

Section: Discussionmentioning

confidence: 99%

Optimal index for detecting splenic involvement on 18F-fluorodeoxyglucose positron emission tomography/computed tomography imaging in diffuse large B-cell lymphoma

Kozuki,

Sugimoto,

Goto

et al. 2024

Medicine

View full text Add to dashboard Cite

show abstract

“…Another study used the spleen’s radiomics signature on CT to predict the recurrence of HCC [ 34 ]. Additionally, considering lymphoma, Enke et al developed a radiomics model on CT to differentiate between spleen involvement versus controls, as well as to discriminate different subtypes of malignant lymphoma [ 35 ]. Their results showed that the radiomics signature could predict the presence of malignant lymphoma with an AUC of 0.86, and even differentiate between subtypes with a satisfying AUC.…”

Section: Discussionmentioning

confidence: 99%

Two-Stage Deep Learning Model for Automated Segmentation and Classification of Splenomegaly

Meddeb

Kossen

Bressem

et al. 2022

Cancers

View full text Add to dashboard Cite

Splenomegaly is a common cross-sectional imaging finding with a variety of differential diagnoses. This study aimed to evaluate whether a deep learning model could automatically segment the spleen and identify the cause of splenomegaly in patients with cirrhotic portal hypertension versus patients with lymphoma disease. This retrospective study included 149 patients with splenomegaly on computed tomography (CT) images (77 patients with cirrhotic portal hypertension, 72 patients with lymphoma) who underwent a CT scan between October 2020 and July 2021. The dataset was divided into a training (n = 99), a validation (n = 25) and a test cohort (n = 25). In the first stage, the spleen was automatically segmented using a modified U-Net architecture. In the second stage, the CT images were classified into two groups using a 3D DenseNet to discriminate between the causes of splenomegaly, first using the whole abdominal CT, and second using only the spleen segmentation mask. The classification performances were evaluated using the area under the receiver operating characteristic curve (AUC), accuracy (ACC), sensitivity (SEN), and specificity (SPE). Occlusion sensitivity maps were applied to the whole abdominal CT images, to illustrate which regions were important for the prediction. When trained on the whole abdominal CT volume, the DenseNet was able to differentiate between the lymphoma and liver cirrhosis in the test cohort with an AUC of 0.88 and an ACC of 0.88. When the model was trained on the spleen segmentation mask, the performance decreased (AUC = 0.81, ACC = 0.76). Our model was able to accurately segment splenomegaly and recognize the underlying cause. Training on whole abdomen scans outperformed training using the segmentation mask. Nonetheless, considering the performance, a broader and more general application to differentiate other causes for splenomegaly is also conceivable.

show abstract

“…In total, 1316 features were extracted per lesion. The features were not harmonized to account for different scanner types, as this has been shown to not improve model performance in previous work on CT-based features [ 20 ].…”

Section: Methodsmentioning

confidence: 99%

Combination of Whole-Body Baseline CT Radiomics and Clinical Parameters to Predict Response and Survival in a Stage-IV Melanoma Cohort Undergoing Immunotherapy

Peisen

Hänsch

Hering

et al. 2022

Cancers

Self Cite

View full text Add to dashboard Cite

Background: This study investigated whether a machine-learning-based combination of radiomics and clinical parameters was superior to the use of clinical parameters alone in predicting therapy response after three months, and overall survival after six and twelve months, in stage-IV malignant melanoma patients undergoing immunotherapy with PD-1 checkpoint inhibitors and CTLA-4 checkpoint inhibitors. Methods: A random forest model using clinical parameters (demographic variables and tumor markers = baseline model) was compared to a random forest model using clinical parameters and radiomics (extended model) via repeated 5-fold cross-validation. For this purpose, the baseline computed tomographies of 262 stage-IV malignant melanoma patients treated at a tertiary referral center were identified in the Central Malignant Melanoma Registry, and all visible metastases were three-dimensionally segmented (n = 6404). Results: The extended model was not significantly superior compared to the baseline model for survival prediction after six and twelve months (AUC (95% CI): 0.664 (0.598, 0.729) vs. 0.620 (0.545, 0.692) and AUC (95% CI): 0.600 (0.526, 0.667) vs. 0.588 (0.481, 0.629), respectively). The extended model was not significantly superior compared to the baseline model for response prediction after three months (AUC (95% CI): 0.641 (0.581, 0.700) vs. 0.656 (0.587, 0.719)). Conclusion: The study indicated a potential, but non-significant, added value of radiomics for six-month and twelve-month survival prediction of stage-IV melanoma patients undergoing immunotherapy.

show abstract

Radiomics Features of the Spleen as Surrogates for CT-Based Lymphoma Diagnosis and Subtype Differentiation

Cited by 16 publications

References 44 publications

Optimal index for detecting splenic involvement on 18F-fluorodeoxyglucose positron emission tomography/computed tomography imaging in diffuse large B-cell lymphoma

Optimal index for detecting splenic involvement on 18F-fluorodeoxyglucose positron emission tomography/computed tomography imaging in diffuse large B-cell lymphoma

Two-Stage Deep Learning Model for Automated Segmentation and Classification of Splenomegaly

Combination of Whole-Body Baseline CT Radiomics and Clinical Parameters to Predict Response and Survival in a Stage-IV Melanoma Cohort Undergoing Immunotherapy

Contact Info

Product

Resources

About