Clinical application of AI-based PET images in oncological patients

Remote Control of Energy Transformation-Based Cancer Imaging and Therapy

Cancer treatment requires precise tumor-specific targeting at specific sites that allows for high-resolution diagnostic imaging and long-term patient-tailorable cancer therapy; while, minimizing side effects largely arising from non-targetability. This can be realized by harnessing exogenous remote stimuli, such as tissue-penetrative ultrasound, magnetic field, light, and radiation, that enable local activation for cancer imaging and therapy in deep tumors. A myriad of nanomedicines can be efficiently activated when the energy of such remote stimuli can be transformed into another type of energy. This review discusses the remote control of energy transformation for targetable, efficient, and long-term cancer imaging and therapy. Such ultrasonic, magnetic, photonic, radiative, and radioactive energy can be transformed into mechanical, thermal, chemical, and radiative energy to enable a variety of cancer imaging and treatment modalities. The current review article describes multimodal energy transformation where a serial cascade or multiple types of energy transformation occur. This review includes not only mechanical, chemical, hyperthermia, and radiation therapy but also emerging thermoelectric, pyroelectric, and piezoelectric therapies for cancer treatment. It also illustrates ultrasound, magnetic resonance, fluorescence, computed tomography, photoluminescence, and photoacoustic imaging-guided cancer therapies. It highlights afterglow imaging that can eliminate autofluorescence for sustained signal emission after the excitation.

Rapid genomic sequencing for genetic disease diagnosis and therapy in intensive care units: a review

Single locus (Mendelian) diseases are a leading cause of childhood hospitalization, intensive care unit (ICU) admission, mortality, and healthcare cost. Rapid genome sequencing (RGS), ultra-rapid genome sequencing (URGS), and rapid exome sequencing (RES) are diagnostic tests for genetic diseases for ICU patients. In 44 studies of children in ICUs with diseases of unknown etiology, 37% received a genetic diagnosis, 26% had consequent changes in management, and net healthcare costs were reduced by $14,265 per child tested by URGS, RGS, or RES. URGS outperformed RGS and RES with faster time to diagnosis, and higher rate of diagnosis and clinical utility. Diagnostic and clinical outcomes will improve as methods evolve, costs decrease, and testing is implemented within precision medicine delivery systems attuned to ICU needs. URGS, RGS, and RES are currently performed in <5% of the ~200,000 children likely to benefit annually due to lack of payor coverage, inadequate reimbursement, hospital policies, hospitalist unfamiliarity, under-recognition of possible genetic diseases, and current formatting as tests rather than as a rapid precision medicine delivery system. The gap between actual and optimal outcomes in children in ICUs is currently increasing since expanded use of URGS, RGS, and RES lags growth in those likely to benefit through new therapies. There is sufficient evidence to conclude that URGS, RGS, or RES should be considered in all children with diseases of uncertain etiology at ICU admission. Minimally, diagnostic URGS, RGS, or RES should be ordered early during admissions of critically ill infants and children with suspected genetic diseases.

18F-Fluoroethyl-L Tyrosine Positron Emission Tomography Radiomics in the Differentiation of Treatment-Related Changes from Disease Progression in Patients with Glioblastoma

The follow-up of glioma patients after therapeutic intervention remains a challenging topic, as therapy-related changes can emulate true progression in contrast-enhanced magnetic resonance imaging. 18F-fluoroethyl-tyrosine (18F-FET) is a radiopharmaceutical that accumulates in glioma cells due to an increased expression of L-amino acid transporters and, contrary to gadolinium, does not depend on blood-brain barrier disruption to reach tumoral cells. It has demonstrated a high diagnostic value in the differentiation of tumoral viability and pseudoprogression or any other therapy-related changes, especially when combining traditional visual analysis with modern radiomics. In this review, we aim to cover the potential role of 18F-FET positron emission tomography in everyday clinical practice when applied to the follow-up of patients after the first therapeutical intervention, early response evaluation, and the differential diagnosis between therapy-related changes and progression.

Improving the efficacy of anti-EGFR drugs in GBM: Where we are going?

The therapies targeting mutations of driver genes in cancer have advanced into clinical trials for a variety of tumors. In glioblastoma (GBM), epidermal growth factor receptor (EGFR) is the most commonly mutated oncogene, and targeting EGFR has been widely investigated as a promising direction. However, the results of EGFR pathway inhibitors have not been satisfactory. Limited blood-brain barrier (BBB) permeability, drug resistance, and pathway compensation mechanisms contribute to the failure of anti-EGFR therapies. This review summarizes recent research advances in EGFR-targeted therapy for GBM and provides insight into the reasons for the unsatisfactory results of EGFR-targeted therapy. By combining the results of preclinical studies with those of clinical trials, we discuss that improved drug penetration across the BBB, the use of multi-target combinations, and the development of peptidomimetic drugs under the premise of precision medicine may be promising strategies to overcome drug resistance in GBM.

A Machine Learning Approach Using [18F]FDG PET-Based Radiomics for Prediction of Tumor Grade and Prognosis in Pancreatic Neuroendocrine Tumor

PURPOSE

We sought to develop and validate machine learning (ML) models for predicting tumor grade and prognosis using 2-[18F]fluoro-2-deoxy-D-glucose ([18F]FDG) positron emission tomography (PET)-based radiomics and clinical features in patients with pancreatic neuroendocrine tumors (PNETs).

PROCEDURES

A total of 58 patients with PNETs who underwent pretherapeutic [18F]FDG PET/computed tomography (CT) were retrospectively enrolled. PET-based radiomics extracted from segmented tumor and clinical features were selected to develop prediction models by the least absolute shrinkage and selection operator feature selection method. The predictive performances of ML models using neural network (NN) and random forest algorithms were compared by the areas under the receiver operating characteristic curves (AUROCs) and validated by stratified five-fold cross validation.

RESULTS

We developed two separate ML models for predicting high-grade tumors (Grade 3) and tumors with poor prognosis (disease progression within two years). The integrated models consisting of clinical and radiomic features with NN algorithm showed the best performances than the other models (stand-alone clinical or radiomics models). The performance metrics of the integrated model by NN algorithm were AUROC of 0.864 in the tumor grade prediction model and AUROC of 0.830 in the prognosis prediction model. In addition, AUROC of the integrated clinico-radiomics model with NN was significantly higher than that of tumor maximum standardized uptake model in predicting prognosis (P < 0.001).

CONCLUSIONS

Integration of clinical features and [18F]FDG PET-based radiomics using ML algorithms improved the prediction of high-grade PNET and poor prognosis in a non-invasive manner.

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.