Prediction of early metastatic disease in experimental breast cancer bone metastasis by combining PET/CT and MRI parameters to a Model-Averaged Neural Network

Development of a nomogram to predict the prognosis of patients with secondary bone tumors in the intensive care unit: a retrospective analysis based on the MIMIC IV database

PURPOSE

The present study aimed to develop a nomogram to predict the prognosis of patients with secondary bone tumors in the intensive care unit to facilitate risk stratification and treatment planning.

METHODS

We used the MIMIC IV 2.0 (the Medical Information Mart for Intensive Care IV) to retrieve patients with secondary bone tumors as a study cohort. To evaluate the predictive ability of each characteristic on patient mortality, stepwise Cox regression was used to screen variables, and the selected variables were included in the final Cox proportional hazard model. Finally, the performance of the model was tested using the decision curve, calibration curve, and receiver operating characteristic (ROC) curve.

RESULTS

A total of 1028 patients were enrolled after excluding cases with missing information. In the training cohort, albumin, APSIII (Acute Physiology Score III), chemotherapy, lactate, chloride, hepatic metastases, respiratory failure, SAPSII (Simplified Acute Physiology Score II), and total protein were identified as independent risk factors for patient death and then incorporated into the final model. The model showed good and robust prediction performance.

CONCLUSION

We developed a nomogram prognostic model for patients with secondary bone tumors in the intensive care unit, which provides effective survival prediction information.

Artificial intelligence-based radiomics in bone tumors: Technical advances and clinical application

Radiomics is the extraction of predefined mathematic features from medical images for predicting variables of clinical interest. Recent research has demonstrated that radiomics can be processed by artificial intelligence algorithms to reveal complex patterns and trends for diagnosis, and prediction of prognosis and response to treatment modalities in various types of cancer. Artificial intelligence tools can utilize radiological images to solve next-generation issues in clinical decision making. Bone tumors can be classified as primary and secondary (metastatic) tumors. Osteosarcoma, Ewing sarcoma, and chondrosarcoma are the dominating primary tumors of bone. The development of bone tumor model systems and relevant research, and the assessment of novel treatment methods are ongoing to improve clinical outcomes, notably for patients with metastases. Artificial intelligence and radiomics have been utilized in almost full spectrum of clinical care of bone tumors. Radiomics models have achieved excellent performance in the diagnosis and grading of bone tumors. Furthermore, the models enable to predict overall survival, metastases, and recurrence. Radiomics features have exhibited promise in assisting therapeutic planning and evaluation, especially neoadjuvant chemotherapy. This review provides an overview of the evolution and opportunities for artificial intelligence in imaging, with a focus on hand-crafted features and deep learning-based radiomics approaches. We summarize the current application of artificial intelligence-based radiomics both in primary and metastatic bone tumors, and discuss the limitations and future opportunities of artificial intelligence-based radiomics in this field. In the era of personalized medicine, our in-depth understanding of emerging artificial intelligence-based radiomics approaches will bring innovative solutions to bone tumors and achieve clinical application.

Development and validation of a risk prediction model for cage subsidence after instrumented posterior lumbar fusion based on machine learning: a retrospective observational cohort study

Background

Interbody cage subsidence is a common complication after instrumented posterior lumbar fusion surgery, several previous studies have shown that cage subsidence is related to multiple factors. But the current research has not combined these factors to predict the subsidence, there is a lack of an individualized and comprehensive evaluation of the risk of cage subsidence following the surgery. So we attempt to identify potential risk factors and develop a risk prediction model that can predict the possibility of subsidence by providing a Cage Subsidence Score (CSS) after surgery, and evaluate whether machine learning-related techniques can effectively predict the subsidence.

Methods

This study reviewed 59 patients who underwent posterior lumbar fusion in our hospital from 2014 to 2019. They were divided into a subsidence group and a non-subsidence group according to whether the interbody fusion cage subsidence occurred during follow-up. Data were collected on the patient, including age, sex, cage segment, number of fusion segments, preoperative space height, postoperative space height, preoperative L4 lordosis Angle, postoperative L4 lordosis Angle, preoperative L5 lordosis Angle, postoperative PT, postoperative SS, postoperative PI. The conventional statistical analysis method was used to find potential risk factors that can lead to subsidence, then the results were incorporated into stepwise regression and machine learning algorithms, respectively, to build a model that could predict the subsidence. Finally the diagnostic efficiency of prediction is verified.

Results

Univariate analysis showed significant differences in pre-/postoperative intervertebral disc height, postoperative L4 segment lordosis, postoperative PT, and postoperative SS between the subsidence group and the non-subsidence group (p < 0.05). The CSS was trained by stepwise regression: 2 points for postoperative disc height > 14.68 mm, 3 points for postoperative L4 segment lordosis angle >16.91°, and 4 points for postoperative PT > 22.69°. If the total score is larger than 0.5, it is the high-risk subsidence group, while less than 0.5 is low-risk. The score obtains the area under the curve (AUC) of 0.857 and 0.806 in the development and validation set, respectively. The AUC of the GBM model based on the machine learning algorithm to predict the risk in the training set is 0.971 and the validation set is 0.889. The AUC of the avNNet model reached 0.931 in the training set and 0.868 in the validation set, respectively.

Conclusion

The machine learning algorithm has advantages in some indicators, and we have preliminarily established a CSS that can predict the risk of postoperative subsidence after lumbar fusion and confirmed the important application prospect of machine learning in solving practical clinical problems.

A Cohort Study to Evaluate the Efficacy and Value of CT Perfusion Imaging in Patients with Metastatic Osteosarcoma after Chemotherapy

Objective

A case-control study was conducted to explore the efficacy of cohort study and value of CT perfusion imaging in patients with metastatic osteosarcoma after chemotherapy.

Methods

Eighty patients with metastatic osteosarcoma treated in our hospital from March 2020 to December 2021 were divided into two groups. According to their different treatment methods, the chemotherapy+antiangiogenesis group had 36 cases and the chemotherapy group had 44 cases. All patients were scanned by 64-slice spiral CT before and after treatment. The differences of tumor volume and perfusion parameters before and after treatment were compared, and the correlation between perfusion parameters and tumor microvessel density (MVD) was analyzed. The receiver working curve (ROC curve) was used to evaluate the efficacy of the two groups after chemotherapy.

Results

Blood flow (BF), blood volume (BV), Pallak blood volume (PBV), and time to start (TTS) in the antitumor angiogenesis+chemotherapy group were significantly lower than those before treatment (P < 0.05). Microvessel density was positively correlated with PS, BF, BV, and PBV (P < 0.05). The reduction rate of BV and BF in the remission group after treatment was significantly higher than that in the nonremission group. When the BV and BF decline rates were 47.37% and 21.53% and the areas under the curve were 0.968 and 0.916, respectively, the diagnostic effect was the best. When the decrease rate of BV was 47.48% and the decrease rate of BF was 21.55%, the sensitivity was 94.72% and 89.56% and the specificity was 91.31% and 91.31%.

Conclusion

The reduction rate of BV and BF in CT perfusion imaging is of high value in evaluating the efficacy of radiotherapy and chemotherapy in patients with NSCLC and can provide more objective basis for observing the changes and judging the prognosis of osteosarcoma after treatment.

Non-Invasive Characterization of Experimental Bone Metastasis in Obesity Using Multiparametric MRI and PET/CT

The growth of primary tumors and metastases is associated with excess body fat. In bone metastasis formation, the bone marrow microenvironment, and particularly adipocytes, play a pivotal role as growth mediators of disseminated tumor cells in the bone marrow. The aim of the present study is to non-invasively characterize the pathophysiologic processes in experimental bone metastasis resulting from accelerated tumor progression within adipocyte-rich bone marrow using multimodal imaging from magnetic resonance imaging (MRI) and positron emission tomography/computed tomography (PET/CT). To achieve this, we have employed small animal models after the administration of MDA-MB 231 breast cancer and B16F10 melanoma cells into the bone of nude rats or C57BL/6 mice, respectively. After tumor cell inoculation, ultra-high field MRI and µPET/CT were used to assess functional and metabolic parameters in the bone marrow of control animals (normal diet, ND), following a high-fat diet (HFD), and/or treated with the peroxisome proliferator-activated receptor-gamma (PPARγ) antagonist bisphenol-A-diglycidylether (BADGE), respectively. In the bone marrow of nude rats, dynamic contrast-enhanced MRI (DCE-MRI) and diffusion-weighted imaging (DWI), as well as [18F]fluorodeoxyglucose-PET/CT([18F]FDG-PET/CT), was performed 10, 20, and 30 days after tumor cell inoculation, followed by immunohistochemistry. DCE-MRI parameters associated with blood volume, such as area under the curve (AUC), were significantly increased in bone metastases in the HFD group 30 days after tumor cell inoculation as compared to controls (p < 0.05), while the DWI parameter apparent diffusion coefficient (ADC) was not significantly different between the groups. [18F]FDG-PET/CT showed an enhanced glucose metabolism due to increased standardized uptake value (SUV) at day 30 after tumor cell inoculation in animals that received HFD (p < 0.05). BADGE treatment resulted in the inversion of quantitative DCE-MRI and [18F]FDG-PET/CT data, namely a significant decrease in AUC and SUV in HFD-fed animals as compared to ND-fed controls (p < 0.05). Finally, immunohistochemistry and qPCR confirmed the HFD-induced stimulation in vascularization and glucose activity in murine bone metastases. In conclusion, multimodal and multiparametric MRI and [18F]FDG-PET/CT were able to derive quantitative parameters in bone metastases, revealing an increase in vascularization and glucose metabolism following HFD. Thus, non-invasive imaging may serve as a biomarker for assessing the pathophysiology of bone metastasis in obesity, opening novel options for therapy and treatment monitoring by MRI and [18F]FDG-PET/CT.

Breast Cancer Induced Bone Osteolysis Prediction Using Temporal Variational Autoencoders

Objective and Impact Statement. We adopt a deep learning model for bone osteolysis prediction on computed tomography (CT) images of murine breast cancer bone metastases. Given the bone CT scans at previous time steps, the model incorporates the bone-cancer interactions learned from the sequential images and generates future CT images. Its ability of predicting the development of bone lesions in cancer-invading bones can assist in assessing the risk of impending fractures and choosing proper treatments in breast cancer bone metastasis. Introduction. Breast cancer often metastasizes to bone, causes osteolytic lesions, and results in skeletal-related events (SREs) including severe pain and even fatal fractures. Although current imaging techniques can detect macroscopic bone lesions, predicting the occurrence and progression of bone lesions remains a challenge. Methods. We adopt a temporal variational autoencoder (T-VAE) model that utilizes a combination of variational autoencoders and long short-term memory networks to predict bone lesion emergence on our micro-CT dataset containing sequential images of murine tibiae. Given the CT scans of murine tibiae at early weeks, our model can learn the distribution of their future states from data. Results. We test our model against other deep learning-based prediction models on the bone lesion progression prediction task. Our model produces much more accurate predictions than existing models under various evaluation metrics. Conclusion. We develop a deep learning framework that can accurately predict and visualize the progression of osteolytic bone lesions. It will assist in planning and evaluating treatment strategies to prevent SREs in breast cancer patients.

Systemic PPARγ Antagonism Reduces Metastatic Tumor Progression in Adipocyte-Rich Bone in Excess Weight Male Rodents

Primary tumors are widely associated with an excess in body fat. The role of adipose tissue on tumor cell homing to bone is yet poorly defined. In this study, we aimed to assess whether bone colonization by tumor cells is favored by an adipocyte-rich bone marrow. We delineated the accompanying alterations of the bone microenvironment and established a treatment approach that interferes with high fat diet (HFD)-induced bone metastasis formation. We were able to show that adipocytes affect skeletal tumor growth in a metastatic model of breast cancer in male rats and melanoma in male mice as well as in human breast cancer bone biopsies. Indeed, HFD-induced bone marrow adiposity was accompanied by accelerated tumor progression and increased osteolytic lesions. In human bone metastases, bone marrow adiposity correlated with tumor cell proliferation. By antagonization of the adipocyte differentiation and storage pathway linked to the peroxisome proliferator-activated receptor gamma (PPARγ) with bisphenol-A-diglycidylether (BADGE), we were able to decelerate tumor progression and subsequent osteolytic damage in the bones of two distinct metastatic animal models exposed to HFD. Overall these data show that adipose tissue is a critical factor in bone metastases and cancer-induced bone loss. © 2021 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Bone Metastases Are Measurable: The Role of Whole-Body MRI and Positron Emission Tomography

Metastatic tumor deposits in bone marrow elicit differential bone responses that vary with the type of malignancy. This results in either sclerotic, lytic, or mixed bone lesions, which can change in morphology due to treatment effects and/or secondary bone remodeling. Hence, morphological imaging is regarded unsuitable for response assessment of bone metastases and in the current Response Evaluation Criteria In Solid Tumors 1.1 (RECIST1.1) guideline bone metastases are deemed unmeasurable. Nevertheless, the advent of functional and molecular imaging modalities such as whole-body magnetic resonance imaging (WB-MRI) and positron emission tomography (PET) has improved the ability for follow-up of bone metastases, regardless of their morphology. Both these modalities not only have improved sensitivity for visual detection of bone lesions, but also allow for objective measurements of bone lesion characteristics. WB-MRI provides a global assessment of skeletal metastases and for a one-step "all-organ" approach of metastatic disease. Novel MRI techniques include diffusion-weighted imaging (DWI) targeting highly cellular lesions, dynamic contrast-enhanced MRI (DCE-MRI) for quantitative assessment of bone lesion vascularization, and multiparametric MRI (mpMRI) combining anatomical and functional sequences. Recommendations for a homogenization of MRI image acquisitions and generalizable response criteria have been developed. For PET, many metabolic and molecular radiotracers are available, some targeting tumor characteristics not confined to cancer type (e.g. 18F-FDG) while other targeted radiotracers target specific molecular characteristics, such as prostate specific membrane antigen (PSMA) ligands for prostate cancer. Supporting data on quantitative PET analysis regarding repeatability, reproducibility, and harmonization of PET/CT system performance is available. Bone metastases detected on PET and MRI can be quantitatively assessed using validated methodologies, both on a whole-body and individual lesion basis. Both have the advantage of covering not only bone lesions but visceral and nodal lesions as well. Hybrid imaging, combining PET with MRI, may provide complementary parameters on the morphologic, functional, metabolic and molecular level of bone metastases in one examination. For clinical implementation of measuring bone metastases in response assessment using WB-MRI and PET, current RECIST1.1 guidelines need to be adapted. This review summarizes available data and insights into imaging of bone metastases using MRI and PET.

Applications of artificial intelligence in oncologic 18F-FDG PET/CT imaging: a systematic review

Artificial intelligence (AI) is a growing field of research that is emerging as a promising adjunct to assist physicians in detection and management of patients with cancer. ¹⁸F-FDG PET imaging helps physicians in detection and management of patients with cancer. In this study we discuss the possible applications of AI in ¹⁸F-FDG PET imaging based on the published studies. A systematic literature review was performed in PubMed on early August 2020 to find the relevant studies. A total of 65 studies were available for review against the inclusion criteria which included studies that developed an AI model based on 18F-FDG PET data in cancer to diagnose, differentiate, delineate, stage, assess response to therapy, determine prognosis, or improve image quality. Thirty-two studies met the inclusion criteria and are discussed in this review. The majority of studies are related to lung cancer. Other studied cancers included breast cancer, cervical cancer, head and neck cancer, lymphoma, pancreatic cancer, and sarcoma. All studies were based on human patients except for one which was performed on rats. According to the included studies, machine learning (ML) models can help in detection, differentiation from benign lesions, segmentation, staging, response assessment, and prognosis determination. Despite the potential benefits of AI in cancer imaging and management, the routine implementation of AI-based models and ¹⁸F-FDG PET-derived radiomics in clinical practice is limited at least partially due to lack of standardized, reproducible, generalizable, and precise techniques.

Morphological, functional, and molecular assessment of breast cancer bone metastases by experimental ultrasound techniques compared with magnetic resonance imaging and histological analysis

BACKGROUND

The imaging of bone metastases, which is regularly performed by cross-sectional modalities, is clinically vital when characterizing and staging osseous lesions. In this paper, we aimed to establish a novel methodology using experimental ultrasound (US) techniques to assess the morphological, functional, and molecular features of breast cancer bone metastases in an animal model, compared with magnetic resonance imaging (MRI) and histological analysis.

MATERIALS AND METHODS

Nude rats were implanted intra-arterially with MDA-MB-231 breast cancer cells to induce osteolytic metastasis in their right hind legs. Once tumors had developed, an experimental US technique using automatic 3D scanning and MRI were performed. For assessment of perfusion, functional imaging techniques included contrast-enhanced US (CEUS) and dynamic contrast-enhanced MRI (DCE-MRI). For molecular ultrasound, anti-VEGFR2 conjugated microbubbles were applied and correlated with immunostaining for VEGFR2 expression.

RESULTS

3D US enabled the automatic assessment of osteolytic lesions, including the largest tumor diameters along the x-, y- and z-axes as well as the segmented tumor volumes, without significant differences between US and MRI (p > 0.18). The CEUS and DCE-MRI of osseous lesions showed corresponding results for the parameters peak enhancement, wash-in area under the curve (both, r > 0.5) and wash-in perfusion index (r > 0.3) when differentiating between tumor, necrotic tissue and healthy muscle tissue (all, p < 0.01). Finally, molecular US allowed the non-invasive assessment of increased VEGFR2 expression in skeletal lesions compared with surrounding muscle tissue (p = 0.03), while a control antibody could not discriminate between these tissues (p = 0.44)-a factor which was confirmed by histological analysis.

CONCLUSION

To the best of our knowledge, this is the first report on an imaging protocol for breast cancer bone metastasis using an experimental US scanner. Therefore, we present a novel methodology to characterize these osseous lesions on the morphological, functional, and molecular level in correlation with MRI and histological analysis.

Machine learning for predictive data analytics in medicine: A review illustrated by cardiovascular and nuclear medicine examples

The evidence-based medicine allows the physician to evaluate the risk-benefit ratio of a treatment through setting and data. Risk-based choices can be done by the doctor using different information. With the emergence of new technologies, a large amount of data is recorded offering interesting perspectives with machine learning for predictive data analytics. Machine learning is an ensemble of methods that process data to model a learning problem. Supervised machine learning algorithms consist in using annotated data to construct the model. This category allows to solve prediction data analytics problems. In this paper, we detail the use of supervised machine learning algorithms for predictive data analytics problems in medicine. In the medical field, data can be split into two categories: medical images and other data. For brevity, our review deals with any kind of medical data excluding images. In this article, we offer a discussion around four supervised machine learning approaches: information-based, similarity-based, probability-based and error-based approaches. Each method is illustrated with detailed cardiovascular and nuclear medicine examples. Our review shows that model ensemble (ME) and support vector machine (SVM) methods are the most popular. SVM, ME and artificial neural networks often lead to better results than those given by other algorithms. In the coming years, more studies, more data, more tools and more methods will, for sure, be proposed.

Applications of artificial intelligence and deep learning in molecular imaging and radiotherapy

This brief review summarizes the major applications of artificial intelligence (AI), in particular deep learning approaches, in molecular imaging and radiation therapy research. To this end, the applications of artificial intelligence in five generic fields of molecular imaging and radiation therapy, including PET instrumentation design, PET image reconstruction quantification and segmentation, image denoising (low-dose imaging), radiation dosimetry and computer-aided diagnosis, and outcome prediction are discussed. This review sets out to cover briefly the fundamental concepts of AI and deep learning followed by a presentation of seminal achievements and the challenges facing their adoption in clinical setting.

Computer-Aided Diagnosis in Multiparametric MRI of the Prostate: An Open-Access Online Tool for Lesion Classification with High Accuracy

Computer-aided diagnosis (CADx) approaches could help to objectify reporting on prostate mpMRI, but their use in many cases is hampered due to common-built algorithms that are not publicly available. The aim of this study was to develop an open-access CADx algorithm with high accuracy for classification of suspicious lesions in mpMRI of the prostate. This retrospective study was approved by the local ethics commission, with waiver of informed consent. A total of 124 patients with 195 reported lesions were included. All patients received mpMRI of the prostate between 2014 and 2017, and transrectal ultrasound (TRUS)-guided and targeted biopsy within a time period of 30 days. Histopathology of the biopsy cores served as a standard of reference. Acquired imaging parameters included the size of the lesion, signal intensity (T2w images), diffusion restriction, prostate volume, and several dynamic parameters along with the clinical parameters patient age and serum PSA level. Inter-reader agreement of the imaging parameters was assessed by calculating intraclass correlation coefficients. The dataset was stratified into a train set and test set (156 and 39 lesions in 100 and 24 patients, respectively). Using the above parameters, a CADx based on an Extreme Gradient Boosting algorithm was developed on the train set, and tested on the test set. Performance optimization was focused on maximizing the area under the Receiver Operating Characteristic curve (ROC_AUC). The algorithm was made publicly available on the internet. The CADx reached an ROC_AUC of 0.908 during training, and 0.913 during testing (p = 0.93). Additionally, established rule-in and rule-out criteria allowed classifying 35.8% of the malignant and 49.4% of the benign lesions with error rates of <2%. All imaging parameters featured excellent inter-reader agreement. This study presents an open-access CADx for classification of suspicious lesions in mpMRI of the prostate with high accuracy. Applying the provided rule-in and rule-out criteria might facilitate to further stratify the management of patients at risk.

Dual-targeting liposome modified by glutamic hexapeptide and folic acid for bone metastatic breast cancer

Bone is the most common organ affected by metastatic breast cancer. Targeting delivery of drugs to bone may not only enhance the treatment efficacy, but also reduce the quantity of drug administered. In order to increase the distribution of paclitaxel (PTX) in bone, herein, a novel bone metastasis-targeted glutamic hexapeptide-folic acid (Glu₆-FA) derivative was designed and synthesized as liposome ligand to deliver PTX to bone metastasis effectively. The liposomes were prepared by thin film hydration method and its particle size, zeta potential, encapsulation efficiency, release profile, stability, hemolysis were also characterized. What's more, the anti-tumor effects of PTX-Glu₆-FA-Lip were confirmed by the detection of cell cycle, migration, and further measurement of microtubule stabilization. In addition, the PTX-Glu₆-FA-Lip showed superior targeting ability in vitro and in vivo evaluation as compared to naked PTX, non-coated, singly-modified and co-modified by physical blending liposomes. All the results suggested that Glu₆-FA-modified liposome showed excellent targeting activity to metastatic bone cancer. These findings suggested that Glu₆-FA-Lip was a promising bone metastasis-targeting carrier for the delivery of PTX. This study may therefore be conducive to the field of bone-targeting drugs delivery.

Next-generation imaging of the skeletal system and its blood supply

Bone is organized in a hierarchical 3D architecture. Traditionally, analysis of the skeletal system was based on bone mass assessment by radiographic methods or on the examination of bone structure by 2D histological sections. Advanced imaging technologies and big data analysis now enable the unprecedented examination of bone and provide new insights into its 3D macrostructure and microstructure. These technologies comprise ex vivo and in vivo methods including high-resolution computed tomography (CT), synchrotron-based imaging, X-ray microscopy, ultra-high-field magnetic resonance imaging (MRI), light-sheet fluorescence microscopy, confocal and intravital two-photon imaging. In concert, these techniques have been used to detect and quantify a novel vascular system of trans-cortical vessels in bone. Furthermore, structures such as the lacunar network, which harbours and connects osteocytes, become accessible for 3D imaging and quantification using these methods. Next-generation imaging of the skeletal system and its blood supply are anticipated to contribute to an entirely new understanding of bone tissue composition and function, from macroscale to nanoscale, in health and disease. These insights could provide the basis for early detection and precision-type intervention of bone disorders in the future.