The last few years have been crucial in defining the most appropriate way to quantitatively assess [¹⁸F]FDG PET images in Multiple Myeloma (MM) patients to detect persistent tumor burden. The visual evaluation of images complements the assessment of Measurable Residual Disease (MRD) in bone marrow samples by multiparameter flow cytometry (MFC) or next-generation sequencing (NGS). The aim of this study was to quantify MRD by analyzing quantitative and texture [¹⁸F]FDG PET features. Whole body [¹⁸F]FDG PET of 39 patients with newly diagnosed MM were included in the database, and visually evaluated by experts in nuclear medicine. A segmentation methodology of the skeleton from CT images and an additional manual segmentation tool were proposed, implemented in a software solution including a graphical user interface. Both the compact bone and the spinal canal were removed from the segmentation to obtain only the bone marrow mask. SUV metrics, GLCM, GLRLM, and NGTDM parameters were extracted from the PET images and evaluated by Mann-Whitney U-tests and Spearman ρ rank correlation as valuable features differentiating PET+/PET- and MFC+/MFC- groups. Seven machine learning algorithms were applied for evaluating the classification performance of the extracted features. Quantitative analysis for PET+/PET- differentiating demonstrated to be significant for most of the variables assessed with Mann-Whitney U-test such as Variance, Energy, and Entropy (p-value = 0.001). Moreover, the quantitative analysis with a balanced database evaluated by Mann-Whitney U-test revealed in even better results with 19 features with p-values < 0.001. On the other hand, radiomics analysis for MFC+/MFC- differentiating demonstrated the necessity of combining MFC evaluation with [¹⁸F]FDG PET assessment in the MRD diagnosis. Machine learning algorithms using the image features for the PET+/PET- classification demonstrated high performance metrics but decreasing for the MFC+/MFC- classification. A proof-of-concept for the extraction and evaluation of bone marrow radiomics features of [¹⁸F]FDG PET images was proposed and implemented. The validation showed the possible use of these features for the image-based assessment of MRD.

The combination of visual assessment of whole body [¹⁸F]FDG PET images and evaluation of bone marrow samples by Multiparameter Flow Cytometry (MFC) or Next-Generation Sequencing (NGS) is currently the most common clinical practice for the detection of Measurable Residual Disease (MRD) in Multiple Myeloma (MM) patients. In this study, radiomic features extracted from the bone marrow biopsy locations are analyzed and compared to those extracted from the whole bone marrow in order to study the representativeness of these biopsy locations in the image-based MRD assessment. Whole body [¹⁸F]FDG PET of 39 patients with newly diagnosed MM were included in the database, and visually evaluated by experts in nuclear medicine. A methodology for the segmentation of biopsy sites from PET images, including sternum and posterior iliac crest, and their subsequent quantification is proposed. First, starting from the bone marrow segmentation, a segmentation of the biopsy sites is performed. Then, segmentations are quantified extracting SUV metrics and radiomic features from the [¹⁸F]FDG PET images and are evaluated by Mann–Whitney U-tests as valuable features differentiating PET+/PET− and MFC+ /MFC− groups. Moreover, correlation between whole bone marrow and biopsy sites is studied by Spearman ρ rank. Classification performance of the radiomics features is evaluated applying seven machine learning algorithms. Statistical analyses reveal that some images features are significant in PET+/PET− differentiation, such as SUV_max, Gray Level Non-Uniformity or Entropy, especially with a balanced database where 16 of the features show a p value < 0.001. Correlation analyses between whole bone marrow and biopsy sites results in significant and acceptable coefficients, with 11 of the variables reaching a correlation coefficient greater than 0.7, with a maximum of 0.853. Machine learning algorithms demonstrate high performances in PET+/PET− classification reaching a maximum AUC of 0.974, but not for MFC+/MFC− classification. The results demonstrate the representativeness of sample sites as well as the effectiveness of extracted features (SUV metrics and radiomic features) from the [¹⁸F]FDG PET images in MRD assessment in MM patients.

Non-Hodgkin lymphoma is a heterogeneous group of cancers that triggers bone marrow infiltration in 20–40% of cases. Bone marrow biopsy in combination with a visual assessment of [¹⁸F]FDG PET/CT images is used to assess the marrow status. Despite the potential of both techniques, they still have limitations due to the subjectivity of visual assessment. The present study aims to develop models based on bone marrow uptake in [¹⁸F]FDG PET/CT images at the time of diagnosis to differentiate bone marrow status. For this purpose, a model trained for skeleton segmentation and based on the U-Net architecture is retrained for bone marrow segmentation from CT images. The mask obtained from this segmentation together with the [¹⁸F]FDG PET image is used to extract radiomics features with which 11 machine learning models for marrow status differentiation are trained. The segmentation model yields very satisfactory results with Jaccard and Dice index values of 0.933 and 0.964, respectively. As for the classification models, a maximum F1_score_weighted and F1_score_macro of 0.962 and 0.747, respectively, are achieved. This highlights the potential of these features for bone marrow assessment, laying the foundation for a new clinical decision support system.

Evidence is emerging about the value of textural features as powerful outcome predictors in cancer lesions. The aim of this study is to evaluate the potential of [¹⁸F]FDG PET/CT conventional and textural parameters as survival predictors in patients with stage III and IV oropharyngeal cancer. The database includes 39 patients. Segmentation of the primary lesions was performed. A total of 48 features were extracted, comprising conventional parameters and textural features. A 2-year follow-up period to analyze the Overall Survival (OS) and Relapse-Free Survival (RFS) rates was defined. Kaplan–Meier and Cox proportional hazards regression analyses were computed. Higher TLG (p = 0.001) and Surface (p = 0.001) are significantly related to better OS in Cox regression analysis after multiple-testing correction. Higher GLZLM_ZLNU (p = 0.001) is significantly related to greater relapse rates in RFS Kaplan–Meier analysis after multiple-testing correction. Quantitative [¹⁸F]FDG PET/CT image features, especially the TLG, have been confirmed as predictors of OS and RFS. Textural features, such as GLZLM_ZLNU, demonstrated a potential predictive value for the OS and RFS of the patients. RFS analysis suggest stabilization of patients adhering to the treatment, showing no relapse events after 20 months of follow-up. [¹⁸F]FDG PET/CT is a useful tool for predicting prognosis after chemoradiation therapy of oropharyngeal cancer patients.

Bone metastasis, emerging oncological therapies, and osteoporosis represent some of the distinct clinical contexts which can result in morphological alterations in bone structure. The visual assessment of these changes through anatomical images is considered suboptimal, emphasizing the importance of precise skeletal segmentation as a valuable aid for its evaluation. In the present study, a neural network model for automatic skeleton segmentation from bidimensional computerized tomography (CT) slices is proposed. A total of 77 CT images and their semimanual skeleton segmentation from two acquisition protocols (whole-body and femur-to-head) are used to form a training group and a testing group. Preprocessing of the images includes four main steps: stretcher removal, thresholding, image clipping, and normalization (with two different techniques: interpatient and intrapatient). Subsequently, five different sets are created and arranged in a randomized order for the training phase. A neural network model based on U-Net architecture is implemented with different values of the number of channels in each feature map and number of epochs. The model with the best performance obtains a Jaccard index (IoU) of 0.959 and a Dice index of 0.979. The resultant model demonstrates the potential of deep learning applied in medical images and proving its utility in bone segmentation.

Background Deep learning (DL) models have demonstrated high performance in classifying skin lesions from dermoscopic images. However, the influence of photoexposure-related factors, such as level of exposure due to the anatomical site and skin phototype, on classification performance remains understudied. Investigating these factors is essential not only to understand their potential impact on bias in model predictions, but also to explore their potential role as risk indicators for skin cancer. Objective This study aims to assess the impact of skin phototype and anatomical-site–related photoexposure on the performance of DL models for malignancy detection, with a focus on potential sources of bias. Methods DL models are trained on widely used public dermoscopic image datasets. Performance is then evaluated on a recently published dataset of 60 patients from the University Hospital of North Norway, which includes dermoscopic images and clinical data on skin phototype and anatomical site to assess their impact on model performance. Results Preliminary analysis suggests that model performance varies between subgroups, with reduced precision observed in lesions chronically photoexposed. The reactions that cause red and painful skin are associated with better model performance, in addition to being mostly benign lesions. However, this result is skewed by the low incidence of malignant cases. Conclusions The findings highlight that individual sun-related behaviours and skin characteristics can influence the reliability of DL-based skin lesion identification. These results underscore the importance of evaluating model robustness across diverse patient profiles and may guide future efforts to define healthier sun exposure habits for the population.