Abstract
Purpose. Besides basic measurements as maximum standardized uptake value (SUV)max or SUVmean derived from 18F-FDG positron emission tomography (PET) scans, more advanced quantitative imaging ...features (i.e. "Radiomics" features) are increasingly investigated for treatment monitoring, outcome prediction, or as potential biomarkers. With these prospected applications of Radiomics features, it is a requisite that they provide robust and reliable measurements. The aim of our study was therefore to perform an integrated stability analysis of a large number of PET-derived features in non-small cell lung carcinoma (NSCLC), based on both a test-retest and an inter-observer setup. Methods. Eleven NSCLC patients were included in the test-retest cohort. Patients underwent repeated PET imaging within a one day interval, before any treatment was delivered. Lesions were delineated by applying a threshold of 50% of the maximum uptake value within the tumor. Twenty-three NSCLC patients were included in the inter-observer cohort. Patients underwent a diagnostic whole body PET-computed tomography (CT). Lesions were manually delineated based on fused PET-CT, using a standardized clinical delineation protocol. Delineation was performed independently by five observers, blinded to each other. Fifteen first order statistics, 39 descriptors of intensity volume histograms, eight geometric features and 44 textural features were extracted. For every feature, test-retest and inter-observer stability was assessed with the intra-class correlation coefficient (ICC) and the coefficient of variability, normalized to mean and range. Similarity between test-retest and inter-observer stability rankings of features was assessed with Spearman's rank correlation coefficient. Results. Results showed that the majority of assessed features had both a high test-retest (71%) and inter-observer (91%) stability in terms of their ICC. Overall, features more stable in repeated PET imaging were also found to be more robust against inter-observer variability. Conclusion. Results suggest that further research of quantitative imaging features is warranted with respect to more advanced applications of PET imaging as being used for treatment monitoring, outcome prediction or imaging biomarkers.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, UILJ, UKNU, UL, UM, UPUK
This open access book comprehensively covers the fundamentals of clinical data science, focusing on data collection, modelling and clinical applications. Topics covered in the first section on data ...collection include: data sources, data at scale (big data), data stewardship (FAIR data) and related privacy concerns. Aspects of predictive modelling using techniques such as classification, regression or clustering, and prediction model validation will be covered in the second section. The third section covers aspects of (mobile) clinical decision support systems, operational excellence and value-based healthcare. Fundamentals of Clinical Data Science is an essential resource for healthcare professionals and IT consultants intending to develop and refine their skills in personalized medicine, using solutions based on large datasets from electronic health records or telemonitoring programmes. The book’s promise is “no math, no code”and will explain the topics in a style that is optimized for a healthcare audience.
Background
In the era of datafication, it is important that medical data are accurate and structured for multiple applications. Especially data for oncological staging need to be accurate to stage ...and treat a patient, as well as population-level surveillance and outcome assessment. To support data extraction from free-text radiological reports, Dutch natural language processing (NLP) algorithm was built to quantify T-stage of pulmonary tumors according to the tumor node metastasis (TNM) classification. This structuring tool was translated and validated on English radiological free-text reports. A rule-based algorithm to classify T-stage was trained and validated on, respectively, 200 and 225 English free-text radiological reports from diagnostic computed tomography (CT) obtained for staging of patients with lung cancer. The automated T-stage extracted by the algorithm from the report was compared to manual staging. A graphical user interface was built for training purposes to visualize the results of the algorithm by highlighting the extracted concepts and its modifying context.
Results
Accuracy of the T-stage classifier was 0.89 in the validation set, 0.84 when considering the T-substages, and 0.76 when only considering tumor size. Results were comparable with the Dutch results (respectively, 0.88, 0.89 and 0.79). Most errors were made due to ambiguity issues that could not be solved by the rule-based nature of the algorithm.
Conclusions
NLP can be successfully applied for staging lung cancer from free-text radiological reports in different languages. Focused introduction of machine learning should be introduced in a hybrid approach to improve performance.
Reports are the standard way of communication between the radiologist and the referring clinician. Efforts are made to improve this communication by, for instance, introducing standardization and ...structured reporting. Natural Language Processing (NLP) is another promising tool which can improve and enhance the radiological report by processing free text. NLP as such adds structure to the report and exposes the information, which in turn can be used for further analysis. This paper describes pre-processing and processing steps and highlights important challenges to overcome in order to successfully implement a free text mining algorithm using NLP tools and machine learning in a small language area, like Dutch. A rule-based algorithm was constructed to classify T-stage of pulmonary oncology from the original free text radiological report, based on the items tumor size, presence and involvement according to the 8th TNM classification system. PyContextNLP, spaCy and regular expressions were used as tools to extract the correct information and process the free text. Overall accuracy of the algorithm for evaluating T-stage was 0,83 in the training set and 0,87 in the validation set, which shows that the approach in this pilot study is promising. Future research with larger datasets and external validation is needed to be able to introduce more machine learning approaches and perhaps to reduce required input efforts of domain-specific knowledge. However, a hybrid NLP approach will probably achieve the best results.
Our hypothesis was that pretreatment inflammation in the lung makes pulmonary tissue more susceptible to radiation damage. The relationship between pretreatment (18)Ffluorodeoxyglucose ((18)FFDG) ...uptake in the lungs (as a surrogate for inflammation) and the delivered radiation dose and radiation-induced lung toxicity (RILT) was investigated.
We retrospectively studied a prospectively obtained cohort of 101 non-small-cell lung cancer patients treated with (chemo)radiation therapy (RT). (18)FFDG-positron emission tomography-computed tomography (PET-CT) scans used for treatment planning were studied. Different parameters were used to describe (18)FFDG uptake patterns in the lungs, excluding clinical target volumes, and the interaction with radiation dose. An increase in the dyspnea grade of 1 (Common Terminology Criteria for Adverse Events version 3.0) or more points compared to the pre-RT score was used as an endpoint for analysis of RILT. The effect of (18)FFDG and CT-based variables, dose, and other patient or treatment characteristics that effected RILT was studied using logistic regression.
Increased lung density and pretreatment (18)FFDG uptake were related to RILT after RT with univariable logistic regression. The 95th percentile of the (18)FFDG uptake in the lungs remained significant in multivariable logistic regression (p = 0.016; odds ratio OR = 4.3), together with age (p = 0.029; OR = 1.06), and a pre-RT dyspnea score of ≥1 (p = 0.005; OR = 0.20). Significant interaction effects were demonstrated among the 80th, 90th, and 95th percentiles and the relative lung volume receiving more than 2 and 5 Gy.
The risk of RILT increased with the 95th percentile of the (18)FFDG uptake in the lungs, excluding clinical tumor volume (OR = 4.3). The effect became more pronounced as the fraction of the 5%, 10%, and 20% highest standardized uptake value voxels that received more than 2 Gy to 5 Gy increased. Therefore, the risk of RILT may be decreased by applying sophisticated radiotherapy techniques to avoid areas in the lung with high (18)FFDG uptake.
Natural language processing (NLP) can be used to process and structure free text, such as (free text) radiological reports. In radiology, it is important that reports are complete and accurate for ...clinical staging of, for instance, pulmonary oncology. A computed tomography (CT) or positron emission tomography (PET)-CT scan is of great importance in tumor staging, and NLP may be of additional value to the radiological report when used in the staging process as it may be able to extract the T and N stage of the 8th tumor–node–metastasis (TNM) classification system. The purpose of this study is to evaluate a new TN algorithm (TN-PET-CT) by adding a layer of metabolic activity to an already existing rule-based NLP algorithm (TN-CT). This new TN-PET-CT algorithm is capable of staging chest CT examinations as well as PET-CT scans. The study design made it possible to perform a subgroup analysis to test the external validation of the prior TN-CT algorithm. For information extraction and matching, pyContextNLP, SpaCy, and regular expressions were used. Overall TN accuracy score of the TN-PET-CT algorithm was 0.73 and 0.62 in the training and validation set (
N
= 63,
N
= 100). The external validation of the TN-CT classifier (
N
= 65) was 0.72. Overall, it is possible to adjust the TN-CT algorithm into a TN-PET-CT algorithm. However, outcomes highly depend on the accuracy of the report, the used vocabulary, and its context to express, for example, uncertainty. This is true for both the adjusted PET-CT algorithm and for the CT algorithm when applied in another hospital.
Abstract Background and purpose To analyse the results of routine EPID measurements for individualised patient dosimetry. Materials and methods Calibrated camera-based EPIDs were used to measure the ...central field dose, which was compared with a dose prediction at the EPID level. For transit dosimetry, dose data were calculated using patient transmission and scatter, and compared with measured values. Furthermore, measured transit dose data were back-projected to an in vivo dose value at 5 cm depth in water ( D5 ) and directly compared with D5 from the treatment planning system. Dose differences per treatment session were calculated by weighting dose values with the number of monitor units per beam. Reported errors were categorised and analysed for approximately 37,500 images from 2511 patients during a period of 24 months. Results Pre-treatment measurements showed a mean dose difference per treatment session of 0.0 ± 1.7% (1 SD). Transfer errors were detected and corrected prior to the first treatment session. An accelerator output variation of about 4% was found between two weekly QC measurements. Patient dosimetry showed mean transit and D5 dose differences of −0.7 ± 5.2% (1 SD) and −0.3 ± 5.6% (1 SD) per treatment session, respectively. Dose differences could be related to set-up errors, organ motion, erroneous density corrections and changes in patient anatomy. Conclusions EPIDs can be used routinely to accurately verify treatment parameter transfer and machine output. By applying transit and in vivo dosimetry, more insight can be obtained with respect to the different error sources influencing dose delivery to a patient.
Abstract Evidence is accumulating that radiotherapy of non-small cell lung cancer patients can be optimized by escalating the tumour dose until the normal tissue tolerances are met. To further ...improve the therapeutic ratio between tumour control probability and the risk of normal tissue complications, we firstly need to exploit inter patient variation. This variation arises, e.g. from differences in tumour shape and size, lung function and genetic factors. Secondly improvement is achieved by taking into account intra- tumour and intra- organ heterogeneity derived from molecular and functional imaging. Additional radiation dose must be delivered to those parts of the tumour that need it the most, e.g. because of increased radio-resistance or reduced therapeutic drug uptake, and away from regions inside the lung that are most prone to complication. As the delivery of these treatments plans is very sensitive for geometrical uncertainties, probabilistic treatment planning is needed to generate robust treatment plans. The administration of these complicated dose distributions requires a quality assurance procedure that can evaluate the treatment delivery and, if necessary, adapt the treatment plan during radiotherapy.
ObjectiveCardiovascular diseases (CVD) are one of the most prevalent diseases in India amounting for nearly 30% of total deaths. A dearth of research on CVD risk scores in Indian population, limited ...performance of conventional risk scores and inability to reproduce the initial accuracies in randomised clinical trials has led to this study on large-scale patient data. The objective is to develop an Artificial Intelligence-based Risk Score (AICVD) to predict CVD event (eg, acute myocardial infarction/acute coronary syndrome) in the next 10 years and compare the model with the Framingham Heart Risk Score (FHRS) and QRisk3.MethodsOur study included 31 599 participants aged 18–91 years from 2009 to 2018 in six Apollo Hospitals in India. A multistep risk factors selection process using Spearman correlation coefficient and propensity score matching yielded 21 risk factors. A deep learning hazards model was built on risk factors to predict event occurrence (classification) and time to event (hazards model) using multilayered neural network. Further, the model was validated with independent retrospective cohorts of participants from India and the Netherlands and compared with FHRS and QRisk3.ResultsThe deep learning hazards model had a good performance (area under the curve (AUC) 0.853). Validation and comparative results showed AUCs between 0.84 and 0.92 with better positive likelihood ratio (AICVD −6.16 to FHRS −2.24 and QRisk3 −1.16) and accuracy (AICVD −80.15% to FHRS 59.71% and QRisk3 51.57%). In the Netherlands cohort, AICVD also outperformed the Framingham Heart Risk Model (AUC −0.737 vs 0.707).ConclusionsThis study concludes that the novel AI-based CVD Risk Score has a higher predictive performance for cardiac events than conventional risk scores in Indian population.Trial registration numberCTRI/2019/07/020471.
Treatment verification is a prerequisite for the verification of complex treatments, checking both the treatment planning process and the actual beam delivery. Pretreatment verification can detect ...errors introduced by the treatment planning system (TPS) or differences between planned and delivered dose distributions. In a previous paper we described the reconstruction of three-dimensional (3-D) dose distributions in homogeneous phantoms using an in-house developed model based on the beams delivered by the linear accelerator measured with an amorphous silicon electronic portal imaging device (EPID), and a dose calculation engine using the Monte Carlo code XVMC. The aim of the present study is to extend the method to situations in which tissue inhomogeneities are present and to make a comparison with the dose distributions calculated by the TPS. Dose distributions in inhomogeneous phantoms, calculated using the fast-Fourier transform convolution (FFTC) and multigrid superposition (MGS) algorithms present in the TPS, were verified using the EPID-based dose reconstruction method and compared to film and ionization chamber measurements. Differences between dose distributions were evaluated using the
γ
-evaluation method
(
3
%
∕
3
mm
)
and expressed as a mean
γ
and the percentage of points with
γ
>
1
(
P
γ
>
1
)
. For rectangular inhomogeneous phantoms containing a low-density region, the differences between film and reconstructed dose distributions were smaller than 3%. In low-density regions there was an overestimation of the planned dose using the FFTC and MGS algorithms of the TPS up to 20% and 8%, respectively, for a
10
MV
photon beam and a
3
×
3
cm
2
field. For lower energies and larger fields (
6
MV
,
5
×
5
cm
2
), these differences reduced to 6% and 3%, respectively. Dose reconstruction performed in an anthropomorphic thoracic phantom for a 3-D conformal and an IMRT plan, showed good agreement between film data and reconstructed dose values
(
P
γ
>
1
<
6
%
)
. The algorithms of the TPS underestimated the dose in the low-dose regions outside the treatment field, due to an implementation error of the jaws and multileaf collimator of the linac in the TPS. The FFTC algorithm of the TPS showed differences up to 6% or
6
mm
at the interface between lung and breast. Two intensity-modulated radiation therapy head and neck plans, reconstructed in a commercial phantom having a bone-equivalent insert and an air cavity, showed good agreement between film measurement, reconstructed and planned dose distributions using the FFTC and MGS algorithm, except in the bone-equivalent regions where both TPS algorithms underestimated the dose with 4%. Absolute dose verification was performed at the isocenter where both planned and reconstructed dose were within 2% of the measured dose. Reproducibility for the EPID measurements was assessed and found to be of negligible influence on the reconstructed dose distribution. Our 3-D dose verification approach is based on the actual dose measured with an EPID in combination with a Monte Carlo dose engine, and therefore independent of a TPS. Because dose values are reconstructed in 3-D, isodose surfaces and dose-volume histograms can be used to detect dose differences in target volume and normal tissues. Using our method, the combined planning and treatment delivery process is verified, offering an easy to use tool for the verification of complex treatments.