The Evolution of Artificial Intelligence in Medical Imaging: From Computer Science to Machine and Deep Learning.

Artificial intelligence (AI), the wide spectrum of technologies aiming to give machines or computers the ability to perform human-like cognitive functions, began in the 1940s with the first abstract models of intelligent machines. Soon after, in the 1950s and 1960s, machine learning algorithms such as neural networks and decision trees ignited significant enthusiasm. More recent advancements include the refinement of learning algorithms, the development of convolutional neural networks to efficiently analyze images, and methods to synthesize new images.

Deep learning based time-to-event analysis with PET, CT and joint PET/CT for head and neck cancer prognosis.

Objectives: Recent studies have shown that deep learning based on pre-treatment positron emission tomography (PET) or computed tomography (CT) is promising for distant metastasis (DM) and overall survival (OS) prognosis in head and neck cancer (HNC). However, lesion segmentation is typically required, resulting in a predictive power susceptible to variations in primary and lymph node gross tumor volume (GTV) segmentation. This study aimed at achieving prognosis without GTV segmentation, and extending single modality prognosis to joint PET/CT to allow investigating the predictive performance of combined- compared to single-modality inputs.

Combining computed tomography and biologically effective dose in radiomics and deep learning improves prediction of tumor response to robotic lung stereotactic body radiation therapy.

Purpose: The aim of this study is to improve the performance of machine learning (ML) models in predicting response of non-small cell lung cancer (NSCLC) to stereotactic body radiation therapy (SBRT) by integrating image features from pre-treatment computed tomography (CT) with features from the biologically effective dose (BED) distribution.

Materials And Methods: Image features, consisting of crafted radiomic features or machine-learned features extracted using a convolutional neural network, were calculated from pre-treatment CT data and from dose distributions converted into BED for 80 NSCLC lesions over 76 patients treated with robotic guided SBRT. ML models using different combinations of features were trained to predict complete or partial response according to response criteria in solid tumors, including radiomics CT (Rad ), radiomics CT and BED (Rad ), deep learning (DL) CT (DL ), and DL CT and BED (DL ).

Distant metastasis time to event analysis with CNNs in independent head and neck cancer cohorts.

Deep learning models based on medical images play an increasingly important role for cancer outcome prediction. The standard approach involves usage of convolutional neural networks (CNNs) to automatically extract relevant features from the patient's image and perform a binary classification of the occurrence of a given clinical endpoint. In this work, a 2D-CNN and a 3D-CNN for the binary classification of distant metastasis (DM) occurrence in head and neck cancer patients were extended to perform time-to-event analysis.

Machine and deep learning methods for radiomics.

Radiomics is an emerging area in quantitative image analysis that aims to relate large-scale extracted imaging information to clinical and biological endpoints. The development of quantitative imaging methods along with machine learning has enabled the opportunity to move data science research towards translation for more personalized cancer treatments. Accumulating evidence has indeed demonstrated that noninvasive advanced imaging analytics, that is, radiomics, can reveal key components of tumor phenotype for multiple three-dimensional lesions at multiple time points over and beyond the course of treatment.

Electron Density and Biologically Effective Dose (BED) Radiomics-Based Machine Learning Models to Predict Late Radiation-Induced Subcutaneous Fibrosis.

to predict the occurrence of late subcutaneous radiation induced fibrosis (RIF) after partial breast irradiation (PBI) for breast carcinoma by using machine learning (ML) models and radiomic features from 3D Biologically Effective Dose (3D-BED) and Relative Electron Density (3D-RED). 165 patients underwent external PBI following a hypo-fractionation protocol consisting of 40 Gy/10 fractions, 35 Gy/7 fractions, and 28 Gy/4 fractions, for 73, 60, and 32 patients, respectively. Physicians evaluated toxicity at regular intervals by the Common Terminology Adverse Events (CTAE) version 4.

Radiomics and deep learning in lung cancer.

Lung malignancies have been extensively characterized through radiomics and deep learning. By providing a three-dimensional characterization of the lesion, models based on radiomic features from computed tomography (CT) and positron-emission tomography (PET) have been developed to detect nodules, distinguish malignant from benign lesions, characterize their histology, stage, and genotype. Deep learning models have been applied to automatically segment organs at risk in lung cancer radiotherapy, stratify patients according to the risk for local and distant recurrence, and identify patients candidate for molecular targeted therapy and immunotherapy.

Electron radiotherapy (IOERT) for applications outside of the breast: Dosimetry and influence of tissue inhomogeneities.

Purpose: The purpose of study is to investigate the dosimetry of electron intraoperative radiotherapy (IOERT) of the Intraop Mobetron 2000 mobile LINAC in treatments outside of the breast. After commissioning and external validation of dosimetry, we report in vivo results of measurements for treatments outside the breast in a large patient cohort, and investigate if the presence of inhomogeneities can affect in vivo measurements.

Methods And Materials: Applicator factors and profile curves were measured with a stereotactic diode.

AI-based applications in hybrid imaging: how to build smart and truly multi-parametric decision models for radiomics.

Introduction: The quantitative imaging features (radiomics) that can be obtained from the different modalities of current-generation hybrid imaging can give complementary information with regard to the tumour environment, as they measure different morphologic and functional imaging properties. These multi-parametric image descriptors can be combined with artificial intelligence applications into predictive models. It is now the time for hybrid PET/CT and PET/MRI to take the advantage offered by radiomics to assess the added clinical benefit of using multi-parametric models for the personalized diagnosis and prognosis of different disease phenotypes.

Ten daily fractions for partial breast irradiation. Long-term results of a prospective phase II trial.

Partial breast irradiation (PBI) is an effective adjuvant treatment after breast conservative surgery for selected early-stage breast cancer patients. However, the best fractionation scheme is not well defined. Hereby, we report the 5-year clinical outcome and toxicity of a phase II prospective study of a novel regimen to deliver PBI, which consists in 40 Gy delivered in 10 daily fractions.

Prediction of skin dose in low-kV intraoperative radiotherapy using machine learning models trained on results of in vivo dosimetry.

Purpose: The purpose of this study was to implement a machine learning model to predict skin dose from targeted intraoperative (TARGIT) treatment resulting in timely adoption of strategies to limit excessive skin dose.

Methods: A total of 283 patients affected by invasive breast carcinoma underwent TARGIT with a prescribed dose of 6 Gy at 1 cm, after lumpectomy. Radiochromic films were used to measure the dose to the skin for each patient.

Voxel-by-voxel correlation between radiologically radiation induced lung injury and dose after image-guided, intensity modulated radiotherapy for lung tumors.

Purpose: To correlate radiation dose to the risk of severe radiologically-evident radiation-induced lung injury (RRLI) using voxel-by-voxel analysis of the follow-up computed tomography (CT) of patients treated for lung cancer with hypofractionated helical Tomotherapy.

Methods And Materials: The follow-up CT scans from 32 lung cancer patients treated with various regimens (5, 8, and 25 fractions) were registered to pre-treatment CT using deformable image registration (DIR). The change in density was calculated for each voxel within the combined lungs minus the planning target volume (PTV).

Beyond imaging: The promise of radiomics.

The domain of investigation of radiomics consists of large-scale radiological image analysis and association with biological or clinical endpoints. The purpose of the present study is to provide a recent update on the status of this rapidly emerging field by performing a systematic review of the literature on radiomics, with a primary focus on oncologic applications. The systematic literature search, performed in Pubmed using the keywords: "radiomics OR radiomic" provided 97 research papers.

Seven fractions to deliver partial breast irradiation: the toxicity is Low.

Purpose: To assess toxicity and clinical outcome, in breast cancer patients treated with external beam partial breast irradiation (PBI) consisting of 35 Gy in 7 daily fractions (5 Gy/fraction).

Materials And Methods: Patients affected by early-stage breast cancer were enrolled in this phase II trial. Patients had to be 60 years old or over and treated with breast conservative surgery for early stage invasive carcinoma.

Characterization of 3D geometric distortion of magnetic resonance imaging scanners commissioned for radiation therapy planning.

Objective: To develop a method for the assessment and characterization of 3D geometric distortion as part of routine quality assurance for MRI scanners commissioned for Radiation Therapy planning.

Materials And Methods: In this study, the in-plane and through-plane geometric distortions on a 1.5T GE MRI-SIM unit are characterized and the 2D and 3D correction algorithms provided by the vendor are evaluated.

Hypofractionation of partial breast irradiation using radiobiological models.

Purpose: To reduce the fraction number in Partial Breast Irradiation (PBI) with initial prescription of 40 Gy in 10 fractions using radiobiological models with specific focus on risk of moderate/severe radiation-induced fibrosis (RIF) and report clinical results.

Methods And Materials: 68 patients (patient group A) were treated with 40 Gy in 10 fractions delivered by field-in-field, forward-planned IMRT. Isotoxic regimens with decreasing number of fractions were calculated using Biological Effective Dose (BED) to the breast.

Fast time-of-flight camera based surface registration for radiotherapy patient positioning.

Purpose: This work introduces a rigid registration framework for patient positioning in radiotherapy, based on real-time surface acquisition by a time-of-flight (ToF) camera. Dynamic properties of the system are also investigated for future gating/tracking strategies.

Methods: A novel preregistration algorithm, based on translation and rotation-invariant features representing surface structures, was developed.

Complication probability model for subcutaneous fibrosis based on published data of partial and whole breast irradiation.

Purpose: To extend the application of current radiation therapy (RT) based normal tissue complication probability (NTCP) models of radiation-induced fibrosis (RIF) of the breast to include the effects of fractionation, inhomogeneous dose, incomplete recovery, and time after the end of radiotherapy in partial breast irradiation (PBI).

Materials And Methods: An NTCP Lyman model with biologically effective uniform dose (BEUD) with and without a correction for the effect of incomplete repair was used. The time to occurrence of RIF was also taken into account.

Correlation of a hypoxia based tumor control model with observed local control rates in nasopharyngeal carcinoma treated with chemoradiotherapy.

Purpose: To extend the application of current radiation therapy (RT) based tumor control probability (TCP) models of nasopharyngeal carcinoma (NPC) to include the effects of hypoxia and chemoradiotherapy (CRT).

Methods: A TCP model is described based on the linear-quadratic model modified to account for repopulation, chemotherapy, heterogeneity of dose to the tumor, and hypoxia. Sensitivity analysis was performed to determine which parameters exert the greatest influence on the uncertainty of modeled TCP.

Methods for an improved detection of the MRI-CEST effect.

CEST imaging is a recently introduced MRI contrast modality based on the use of endogenous or exogenous molecules whose exchangeable proton pools transfer saturated magnetization to bulk water, thus creating negative contrast. One of the critical issues for further development of these agents is represented by their limited sensitivity in vivo. The aim of this work is to improve the detection of CEST agents by exploring new approaches through which the saturation transfer (ST) effect can be enhanced.

Paramagnetic liposomes as innovative contrast agents for magnetic resonance (MR) molecular imaging applications.

This article illustrates some innovative applications of liposomes loaded with paramagnetic lanthanide-based complexes in MR molecular imaging field. When a relatively high amount of a Gd(III) chelate is encapsulated in the vesicle, the nanosystem can simultaneously affect both the longitudinal (R(1)) and the transverse (R(2)) relaxation rate of the bulk H2O H-atoms, and this finding can be exploited to design improved thermosensitive liposomes whose MRI response is not longer dependent on the concentration of the probe. The observation that the liposome compartmentalization of a paramagnetic Ln(III) complex induce a significant R(2) enhancement, primarily caused by magnetic susceptibility effects, prompted us to test the potential of such agents in cell-targeting MR experiments.

3T MRI evaluation of the accuracy of atlas-based subthalamic nucleus identification.

Modulation of the activity of the subthalamic nucleus (STN) using deep brain stimulation (DBS) in patients with advanced Parkinson's disease is the most common procedure performed today by functional neurosurgeons. The STN contours cannot be entirely identified on common 1.5 T images; in particular, the ventromedial border of the STN often blends with the substantia nigra.