Corrigendum: Head and neck cancer treatment outcome prediction: a comparison between machine learning with conventional radiomics features and deep learning radiomics.

[This corrects the article DOI: 10.3389/fmed.2023.1217037.].

Novel ensemble feature selection techniques applied to high-grade gastroenteropancreatic neuroendocrine neoplasms for the prediction of survival.

BACKGROUND AND OBJECTIVE: Determining the most informative features for predicting the overall survival of patients diagnosed with high-grade gastroenteropancreatic neuroendocrine neoplasms is crucial to improve individual treatment plans for patients, as well as the biological understanding of the disease. The main objective of this study is to evaluate the use of modern ensemble feature selection techniques for this purpose with respect to (a) quantitative performance measures such as predictive performance, (b) clinical interpretability, and (c) the effect of integrating prior expert knowledge. METHODS: The Repeated Elastic Net Technique for Feature Selection (RENT) and the User-Guided Bayesian Framework for Feature Selection (UBayFS) are recently developed ensemble feature selectors investigated in this work. Both allow the user to identify informative features in datasets with low sample sizes and focus on model interpretability. While RENT is purely data-driven, UBayFS can integrate expert knowledge a priori in the feature selection process. In this work, we compare both feature selectors on a dataset comprising 63 patients and 110 features from multiple sources, including baseline patient characteristics, baseline blood values, tumor histology, imaging, and treatment information. RESULTS: Our experiments involve data-driven and expert-driven setups, as well as combinations of both. In a five-fold cross-validated experiment without expert knowledge, our results demonstrate that both feature selectors allow accurate predictions: A reduction from 110 to approximately 20 features (around 82%) delivers near-optimal predictive performances with minor variations according to the choice of the feature selector, the predictive model, and the fold. Thereafter, we use findings from clinical literature as a source of expert knowledge. In addition, expert knowledge has a stabilizing effect on the feature set (an increase in stability of approximately 40%), while the impact on predictive performance is limited. CONCLUSIONS: The features WHO Performance Status, Albumin, Platelets, Ki-67, Tumor Morphology, Total MTV, Total TLG, and SUVmax are the most stable and predictive features in our study. Overall, this study demonstrated the practical value of feature selection in medical applications not only to improve quantitative performance but also to deliver potentially new insights to experts.

Head and neck cancer treatment outcome prediction: a comparison between machine learning with conventional radiomics features and deep learning radiomics.

Background: Radiomics can provide in-depth characterization of cancers for treatment outcome prediction. Conventional radiomics rely on extraction of image features within a pre-defined image region of interest (ROI) which are typically fed to a classification algorithm for prediction of a clinical endpoint. Deep learning radiomics allows for a simpler workflow where images can be used directly as input to a convolutional neural network (CNN) with or without a pre-defined ROI. Purpose: The purpose of this study was to evaluate (i) conventional radiomics and (ii) deep learning radiomics for predicting overall survival (OS) and disease-free survival (DFS) for patients with head and neck squamous cell carcinoma (HNSCC) using pre-treatment 18F-fluorodeoxuglucose positron emission tomography (FDG PET) and computed tomography (CT) images. Materials and methods: FDG PET/CT images and clinical data of patients with HNSCC treated with radio(chemo)therapy at Oslo University Hospital (OUS; n = 139) and Maastricht University Medical Center (MAASTRO; n = 99) were collected retrospectively. OUS data was used for model training and initial evaluation. MAASTRO data was used for external testing to assess cross-institutional generalizability. Models trained on clinical and/or conventional radiomics features, with or without feature selection, were compared to CNNs trained on PET/CT images without or with the gross tumor volume (GTV) included. Model performance was measured using accuracy, area under the receiver operating characteristic curve (AUC), Matthew's correlation coefficient (MCC), and the F1 score calculated for both classes separately. Results: CNNs trained directly on images achieved the highest performance on external data for both endpoints. Adding both clinical and radiomics features to these image-based models increased performance further. Conventional radiomics including clinical data could achieve competitive performance. However, feature selection on clinical and radiomics data lead to overfitting and poor cross-institutional generalizability. CNNs without tumor and node contours achieved close to on-par performance with CNNs including contours. Conclusion: High performance and cross-institutional generalizability can be achieved by combining clinical data, radiomics features and medical images together with deep learning models. However, deep learning models trained on images without contours can achieve competitive performance and could see potential use as an initial screening tool for high-risk patients.

Automatic gross tumor segmentation of canine head and neck cancer using deep learning and cross-species transfer learning.

Background: Radiotherapy (RT) is increasingly being used on dogs with spontaneous head and neck cancer (HNC), which account for a large percentage of veterinary patients treated with RT. Accurate definition of the gross tumor volume (GTV) is a vital part of RT planning, ensuring adequate dose coverage of the tumor while limiting the radiation dose to surrounding tissues. Currently the GTV is contoured manually in medical images, which is a time-consuming and challenging task. Purpose: The purpose of this study was to evaluate the applicability of deep learning-based automatic segmentation of the GTV in canine patients with HNC. Materials and methods: Contrast-enhanced computed tomography (CT) images and corresponding manual GTV contours of 36 canine HNC patients and 197 human HNC patients were included. A 3D U-Net convolutional neural network (CNN) was trained to automatically segment the GTV in canine patients using two main approaches: (i) training models from scratch based solely on canine CT images, and (ii) using cross-species transfer learning where models were pretrained on CT images of human patients and then fine-tuned on CT images of canine patients. For the canine patients, automatic segmentations were assessed using the Dice similarity coefficient (Dice), the positive predictive value, the true positive rate, and surface distance metrics, calculated from a four-fold cross-validation strategy where each fold was used as a validation set and test set once in independent model runs. Results: CNN models trained from scratch on canine data or by using transfer learning obtained mean test set Dice scores of 0.55 and 0.52, respectively, indicating acceptable auto-segmentations, similar to the mean Dice performances reported for CT-based automatic segmentation in human HNC studies. Automatic segmentation of nasal cavity tumors appeared particularly promising, resulting in mean test set Dice scores of 0.69 for both approaches. Conclusion: In conclusion, deep learning-based automatic segmentation of the GTV using CNN models based on canine data only or a cross-species transfer learning approach shows promise for future application in RT of canine HNC patients.

Non-linear shrinking of linear model errors.

BACKGROUND: Artificial neural networks (ANNs) can be a powerful tool for spectroscopic data analysis. Their ability to detect and model complex relations in the data may lead to outstanding predictive capabilities, but the predictions themselves are difficult to interpret due to the lack of understanding of the black box ANN models. ANNs and linear methods can be combined by first fitting a linear model to the data followed by a non-linear fitting of the linear model residuals using an ANN. This paper explores the use of residual modelling in high-dimensional data using modern neural network architectures. RESULTS: By combining linear- and ANN modelling, we demonstrate that it is possible to achieve both good model performance while retaining interpretations from the linear part of the model. The proposed residual modelling approach is evaluated on four high-dimensional datasets, representing two regression and two classification problems. Additionally, a demonstration of possible interpretation techniques are included for all datasets. The study concludes that if the modelling problem contains sufficiently complex data (i.e., non-linearities), the residual modelling can in fact improve the performance of a linear model and achieve similar performance as pure ANN models while retaining valuable interpretations for a large proportion of the variance accounted for. SIGNIFICANCE AND NOVELTY: The paper presents a residual modelling scheme using modern neural network architectures. Furthermore, two novel extensions of residual modelling for classification tasks are proposed. The study is seen as a step towards explainable AI, with the aim of making data modelling using artificial neural networks more transparent.

Deep learning-based automatic delineation of anal cancer gross tumour volume: a multimodality comparison of CT, PET and MRI.

BACKGROUND: Accurate target volume delineation is a prerequisite for high-precision radiotherapy. However, manual delineation is resource-demanding and prone to interobserver variation. An automatic delineation approach could potentially save time and increase delineation consistency. In this study, the applicability of deep learning for fully automatic delineation of the gross tumour volume (GTV) in patients with anal squamous cell carcinoma (ASCC) was evaluated for the first time. An extensive comparison of the effects single modality and multimodality combinations of computed tomography (CT), positron emission tomography (PET), and magnetic resonance imaging (MRI) have on automatic delineation quality was conducted. MATERIAL AND METHODS: 18F-fluorodeoxyglucose PET/CT and contrast-enhanced CT (ceCT) images were collected for 86 patients with ASCC. A subset of 36 patients also underwent a study-specific 3T MRI examination including T2- and diffusion-weighted imaging. The resulting two datasets were analysed separately. A two-dimensional U-Net convolutional neural network (CNN) was trained to delineate the GTV in axial image slices based on single or multimodality image input. Manual GTV delineations constituted the ground truth for CNN model training and evaluation. Models were evaluated using the Dice similarity coefficient (Dice) and surface distance metrics computed from five-fold cross-validation. RESULTS: CNN-generated automatic delineations demonstrated good agreement with the ground truth, resulting in mean Dice scores of 0.65-0.76 and 0.74-0.83 for the 86 and 36-patient datasets, respectively. For both datasets, the highest mean Dice scores were obtained using a multimodal combination of PET and ceCT (0.76-0.83). However, models based on single modality ceCT performed comparably well (0.74-0.81). T2W-only models performed acceptably but were somewhat inferior to the PET/ceCT and ceCT-based models. CONCLUSION: CNNs provided high-quality automatic GTV delineations for both single and multimodality image input, indicating that deep learning may prove a versatile tool for target volume delineation in future patients with ASCC.

A comparison of methods for fully automatic segmentation of tumors and involved nodes in PET/CT of head and neck cancers.

Target volume delineation is a vital but time-consuming and challenging part of radiotherapy, where the goal is to deliver sufficient dose to the target while reducing risks of side effects. For head and neck cancer (HNC) this is complicated by the complex anatomy of the head and neck region and the proximity of target volumes to organs at risk. The purpose of this study was to compare and evaluate conventional PET thresholding methods, six classical machine learning algorithms and a 2D U-Net convolutional neural network (CNN) for automatic gross tumor volume (GTV) segmentation of HNC in PET/CT images. For the latter two approaches the impact of single versus multimodality input on segmentation quality was also assessed. 197 patients were included in the study. The cohort was split into training and test sets (157 and 40 patients, respectively). Five-fold cross-validation was used on the training set for model comparison and selection. Manual GTV delineations represented the ground truth. Tresholding, classical machine learning and CNN segmentation models were ranked separately according to the cross-validation Sørensen-Dice similarity coefficient (Dice). PET thresholding gave a maximum mean Dice of 0.62, whereas classical machine learning resulted in maximum mean Dice scores of 0.24 (CT) and 0.66 (PET; PET/CT). CNN models obtained maximum mean Dice scores of 0.66 (CT), 0.68 (PET) and 0.74 (PET/CT). The difference in cross-validation Dice between multimodality PET/CT and single modality CNN models was significant (p ≤ 0.0001). The top-ranked PET/CT-based CNN model outperformed the best-performing thresholding and classical machine learning models, giving significantly better segmentations in terms of cross-validation and test set Dice, true positive rate, positive predictive value and surface distance-based metrics (p ≤ 0.0001). Thus, deep learning based on multimodality PET/CT input resulted in superior target coverage and less inclusion of surrounding normal tissue.

Deep learning-based auto-delineation of gross tumour volumes and involved nodes in PET/CT images of head and neck cancer patients.

PURPOSE: Identification and delineation of the gross tumour and malignant nodal volume (GTV) in medical images are vital in radiotherapy. We assessed the applicability of convolutional neural networks (CNNs) for fully automatic delineation of the GTV from FDG-PET/CT images of patients with head and neck cancer (HNC). CNN models were compared to manual GTV delineations made by experienced specialists. New structure-based performance metrics were introduced to enable in-depth assessment of auto-delineation of multiple malignant structures in individual patients. METHODS: U-Net CNN models were trained and evaluated on images and manual GTV delineations from 197 HNC patients. The dataset was split into training, validation and test cohorts (n= 142, n = 15 and n = 40, respectively). The Dice score, surface distance metrics and the new structure-based metrics were used for model evaluation. Additionally, auto-delineations were manually assessed by an oncologist for 15 randomly selected patients in the test cohort. RESULTS: The mean Dice scores of the auto-delineations were 55%, 69% and 71% for the CT-based, PET-based and PET/CT-based CNN models, respectively. The PET signal was essential for delineating all structures. Models based on PET/CT images identified 86% of the true GTV structures, whereas models built solely on CT images identified only 55% of the true structures. The oncologist reported very high-quality auto-delineations for 14 out of the 15 randomly selected patients. CONCLUSIONS: CNNs provided high-quality auto-delineations for HNC using multimodality PET/CT. The introduced structure-wise evaluation metrics provided valuable information on CNN model strengths and weaknesses for multi-structure auto-delineation.

Prediction of wine sensory properties using mid-infrared spectra of Cabernet Sauvignon and Chardonnay grape berries and wines.

The study determined optimal parameters to four preprocessing techniques for mid-infrared (MIR) spectra of wines and grape berry homogenates and tested MIR's ability to model sensory properties of research Cabernet Sauvignon and Chardonnay wines. Savitsky-Golay (SG) derivative, smoothing points, and polynomial order, and extended multiplicative signal correction (EMSC) polynomial were investigated as preprocessing techniques at 2, 2, 5, and 3 levels, respectively, all in combination. Preprocessed data were analysed with partial least squares regression (PLS) to model the wine sensory data and the regression coefficients of PLS calibration models (R2) were further analysed with multivariate analysis of variance (MANOVA). SG transformations were significant factors from the MANOVA that influenced R2, while EMSC did not. Overall, PLSR models that predicted wine sensory characteristics gave a poor to moderate R2. Consistently predicting wine sensory attributes within a variety and across vintages is challenging, regardless of using grape or wine spectra as predictors.

SO-PLS as an alternative approach for handling multi-dimensionality in modelling different aspects of consumer expectations.

In the development of sensory and consumer science, data are often collected in several blocks responding to different aspects of consumer experience. Sometimes the task of organizing the data and explaining their relation is non-trivial, especially when considering structural (casual) relationship between data sets. In this sense, PLS path modelling (PLS-PM) has been found as a good tool to model such relations, but this approach faces some issues regarding the assumption of uni-dimensionality of consumers' data blocks. Sequential Orthogonalised PLS path modelling (SO-PLS-PM) has been proposed as an alternative approach to handle the multi-dimensionality and to explain the relations between the original data blocks without any preprocessing of the data. This study aims at comparing the efficacy of SO-PLS-PM and PLS-PM (together with splitting blocks into uni-dimensional sub-blocks) for handling multi-dimensionality. Data sets from two satiety perception studies (yoghurt, biscuit) have been used as illustrations. The main novelty of this paper lies in underlining and solving a major, but little studied problem, related to the assumption of one-dimensional blocks in PLS-PM. The findings from the comparisons indicated that the two approaches (PLS-PM and SO-PLS-PM) highlighted the same main trends for the less complex samples (yoghurt samples): liking was the essential driver of satiation perception and portion size selection; while satiation mainly predicted satiety perception. For the more complex data set - from a sensory perspective - (biscuit samples), the relations between data blocks in PLS-PM model was difficult to interpret, whereas they were well explained by SO-PLS-PM. This underlines the ability of SO-PLS-PM to model multi-dimensional data sets without requiring any preprocessing steps.

Rate-All-That-Apply (RATA) comparison of taste profiles for different sweeteners in black tea, chocolate milk, and natural yogurt.

Growing health concerns have increased interest in reducing the consumption of added sugars, which can be achieved by substituting or replacing sugar with sweeteners to maintain sensory intensity and quality. The growing availability of sweeteners has increased the complexity of the perceptual landscape as sweeteners differ in the qualitative, intensity, and temporal properties. A sweetener that can match the perceptual properties of sucrose in different food matrices is likely to have broad applications. In complex foods, sweetness is influenced by the taste interactions with the existing tastants and possible matrix effects that influence release and perception of sweetness. The current study compared the taste properties of three food matrices (black tea, chocolate milk, and natural yogurt) sweetened by sucrose to those sweetened using eight different sweeteners (acesulfame-K, aspartame, erythritol, luo han guo (Mogroside), palatinose (iso-maltulose), stevia (Reb-A), sucralose, and sucrose-allulose mixture) using Rate-All-That-Apply. The sensory properties of each sweetener differed across matrices, with sucrose-allulose mixture, aspartame, erythritol, palatinose, and sucralose having the most similar taste to sucrose across all foods. By contrast, acesulfame-K, stevia, and luo han guo had taste profiles that most varied from sucrose, characterized by side tastes such as bitterness, chemical taste, and a low sweetness. Sweeteners differed most from sucrose when presented in natural yogurt compared to tea and chocolate milk. A food's taste properties can suppress sweetness intensity and promote undesirable side tastes. Taken together, these findings highlight the importance of testing sweeteners in complex foods and help identify sweeteners and sweetener combinations that can replicate the sweetness of sucrose and support sugar reduction. PRACTICAL APPLICATION: Food manufacturers and researchers can refer to the results of the sensory profiles to identify suitable sweeteners substitutes for sucrose in foods with similar taste profiles to those tested. The current article highlights important changes to sweetener sensory properties when presented in different complex foods, and provides an indication of the potential for calorie reduction by substituting sucrose with a range of low or no calorie sweeteners.

Temporal sweetness and side tastes profiles of 16 sweeteners using temporal check-all-that-apply (TCATA).

An increase in consumer awareness around the negative health impacts of consuming excess added sugars has led to a rise in the replacement of sucrose in foods and beverages. This replacement is often through the use of low or no calorie sweeteners to reduce total calories while maintaining sweetness and palatability. There are a wide variety of sweeteners with diverse physical and caloric compositions which can be used at concentrations estimated to be equi-sweet to sucrose in food products. However, many of the available sweeteners are known to differ in their temporal profiles and may have other side and aftertastes alongside sweetness that need to be considered when comparing their suitability as potential sucrose substitutes. The objective of the current study was to profile and compare the temporal sweetness and qualitative differences of 15 sweeteners to sucrose across nutritive saccharide (sucrose, dextrose, fructose, allulose (D-psicose), palatinose (isomaltulose), sucrose-allulose mixture), nutritive polyol (maltitol, erythritol, mannitol, xylitol, sorbitol) and non-nutritive (acesulfame-k, aspartame, sucralose, stevia (rebaudioside-A or rebA), luo han guo (monk fruit extract or mogroside)) groups, at equi-sweet intensity to 10% w/v sucrose based on a previous psychophysical dose-response study. Using the Temporal Check-all-that-Apply (TCATA) method, 20 participants evaluated a set of 17 sweetener samples (including a sucrose duplicate) in triplicates across three 1-h sessions for the occurrence of six attributes (sweetness, bitterness, metallic taste, chemical taste, honey taste and mouth-drying) over a 60 s period. Sucrose was characterised by a rapid sweetness onset (within first 10 s) to peak citation, and subsequent decay of sweetness with minimal side tastes noted. Acesulfame-K, stevia (rebA) and luo han guo had prominent and long-lasting citations of the undesirable bitter, metallic and chemical side tastes and significantly lower sweetness citations. Allulose, erythritol, sorbitol, aspartame and sucralose were perceived to have bitter, metallic and chemical side tastes, but retained a largely sweet taste profile, with longer residual sweetness for aspartame and sucralose. Nutritive sweeteners dextrose, fructose, maltitol, mannitol, sucrose-allulose mixture, palatinose and xylitol had the most similar temporal and qualitative taste profiles when compared to sucrose, in terms of their sweetness onset, peak sweetness citations, sweetness decay, and side taste profiles.

Postoperative wound dehiscence after laparotomy: a useful healthcare quality indicator? A cohort study based on Norwegian hospital administrative data.

OBJECTIVES: Postoperative wound dehiscence (PWD) is a serious complication to laparotomy, leading to higher mortality, readmissions and cost. The aims of the present study are to investigate whether risk adjusted PWD rates could reliably differentiate between Norwegian hospitals, and whether PWD rates were associated with hospital characteristics such as hospital type and laparotomy volume. DESIGN: Observational study using patient administrative data from all Norwegian hospitals, obtained from the Norwegian Patient Registry, for the period 2011-2015, and linked using the unique person identification number. PARTICIPANTS: All patients undergoing laparotomy, aged at least 15 years, with length of stay at least 2 days and no diagnosis code for immunocompromised state or relating to pregnancy, childbirth and puerperium. The final data set comprised 66 925 patients with 78 086 laparotomy episodes from 47 hospitals. OUTCOMES: The outcome was wound dehiscence, identified by the presence of a wound reclosure code, risk adjusted for patient characteristics and operation type. RESULTS: The final data set comprised 1477 wound dehiscences. Crude PWD rates varied from 0% to 5.1% among hospitals, with an overall rate of 1.89%. Three hospitals with statistically significantly higher PWD than average were identified, after case mix adjustment and correction for multiple comparisons. Hospital volume was not associated with PWD rate, except that hospitals with very few laparotomies had lower PWD rates. CONCLUSIONS: Among Norwegian hospitals, there is considerable variation in PWD rate that cannot be explained by operation type, age or comorbidity. This warrants further investigation into possible causes, such as surgical technique, perioperative procedures or handling of complications. The risk adjusted PWD rate after laparotomy is a candidate quality indicator for Norwegian hospitals.

Application of sequential and orthogonalised-partial least squares (SO-PLS) regression to predict sensory properties of Cabernet Sauvignon wines from grape chemical composition.

The current study determined the applicability of sequential and orthogonalised-partial least squares (SO-PLS) regression to relate Cabernet Sauvignon grape chemical composition to the sensory perception of the corresponding wines. Grape samples (n = 25) were harvested at a similar maturity and vinified identically in 2013. Twelve measures using various (bio)chemical methods were made on grapes. Wines were evaluated using descriptive analysis with a trained panel (n = 10) for sensory profiling. Data was analysed globally using SO-PLS for the entire sensory profiles (SO-PLS2), as well as for single sensory attributes (SO-PLS1). SO-PLS1 models were superior in validated explained variances than SO-PLS2. SO-PLS provided a structured approach in the selection of predictor chemical data sets that best contributed to the correlation of important sensory attributes. This new approach presents great potential for application in other explorative metabolomics studies of food and beverages to address factors such as quality and regional influences.

30-Day Survival Probabilities as a Quality Indicator for Norwegian Hospitals: Data Management and Analysis.

BACKGROUND: The Norwegian Knowledge Centre for the Health Services (NOKC) reports 30-day survival as a quality indicator for Norwegian hospitals. The indicators have been published annually since 2011 on the website of the Norwegian Directorate of Health (www.helsenorge.no), as part of the Norwegian Quality Indicator System authorized by the Ministry of Health. Openness regarding calculation of quality indicators is important, as it provides the opportunity to critically review and discuss the method. The purpose of this article is to describe the data collection, data pre-processing, and data analyses, as carried out by NOKC, for the calculation of 30-day risk-adjusted survival probability as a quality indicator. METHODS AND FINDINGS: Three diagnosis-specific 30-day survival indicators (first time acute myocardial infarction (AMI), stroke and hip fracture) are estimated based on all-cause deaths, occurring in-hospital or out-of-hospital, within 30 days counting from the first day of hospitalization. Furthermore, a hospital-wide (i.e. overall) 30-day survival indicator is calculated. Patient administrative data from all Norwegian hospitals and information from the Norwegian Population Register are retrieved annually, and linked to datasets for previous years. The outcome (alive/death within 30 days) is attributed to every hospital by the fraction of time spent in each hospital. A logistic regression followed by a hierarchical Bayesian analysis is used for the estimation of risk-adjusted survival probabilities. A multiple testing procedure with a false discovery rate of 5% is used to identify hospitals, hospital trusts and regional health authorities with significantly higher/lower survival than the reference. In addition, estimated risk-adjusted survival probabilities are published per hospital, hospital trust and regional health authority. The variation in risk-adjusted survival probabilities across hospitals for AMI shows a decreasing trend over time: estimated survival probabilities for AMI in 2011 varied from 80.6% (in the hospital with lowest estimated survival) to 91.7% (in the hospital with highest estimated survival), whereas it ranged from 83.8% to 91.2% in 2013. CONCLUSIONS: Since 2011, several hospitals and hospital trusts have initiated quality improvement projects, and some of the hospitals have improved the survival over these years. Public reporting of survival/mortality indicators are increasingly being used as quality measures of health care systems. Openness regarding the methods used to calculate the indicators are important, as it provides the opportunity of critically reviewing and discussing the methods in the literature. In this way, the methods employed for establishing the indicators may be improved.

Radiology and fine art.

OBJECTIVE: The radiologic aesthetics of some body parts and internal organs have inspired certain artists to create specific works of art. Our aim was to describe the link between radiology and fine art. We explored 13,625 artworks in the literature produced by 2049 artists and found several thousand photographs in an online image search. The examination revealed 271 radiologic artworks (1.99%) created by 59 artists (2.88%) who mainly applied radiography, sonography, CT, and MRI. CONCLUSION: Some authors produced radiologic artistic photographs, and others used radiologic images to create artful compositions, specific sculptures, or digital works. Many radiologic artworks have symbolic, metaphoric, or conceptual connotations. Radiology is clearly becoming an original and important field of modern art.

Neuroradiology and art: a review and personal contribution.

Radiology has attracted the world of art with the esthetic value of its images, and as a new medium for the artistic expression. In order to investigate the links between neuroradiology and art, we examined 12,763 artworks presented in corresponding publications and in Google images on the Internet. The selected artworks were created by 1,964 authors. To give our own contribution to this field, we produced several artful radiological images using the X-ray of 4 cerebral hemispheres, one dissected brain, serial sections of one head and brain, the vascular casts of 2 brains, the magnetic resonance imaging (MRI) scans of one volunteer, and various options in Photoshop. Among the examined artworks, neuroradiological images were used in 129 artworks (1.01%) that were created by 31 artists (1.58%). The artists applied different radiological techniques: X-ray, angiography, computed tomography (CT), multislice CT, MRI, functional MRI, positron emission tomography (PET) or single-photon emission computed tomography (SPECT), either alone or in various combinations. They used the original images, i.e. radiographs or scans, or their electronic modifications in Photoshop or three-dimensional (3D) software. Some artworks presented the skull, yet others the brain, and still others both, either with or without a head image. The neuroradiological artworks were created as paintings, photographs, digital works and sculptures. Their authors were professional artists, designers, amateurs and radiologists. In conclusion, thanks to the esthetics of some radiological images and the valuable creations of certain artists, neuroradiology has become an important field of contemporary art.

Characterization of selected South African young cultivar wines using FTMIR spectroscopy, gas chromatography, and multivariate data analysis.

The powerful combination of analytical chemistry and chemometrics and its application to wine analysis provide a way to gain knowledge and insight into the inherent chemical composition of wine and to objectively distinguish between wines. Extensive research programs are focused on the chemical characterization of wine to establish industry benchmarks and authentication systems. The aim of this study was to investigate the volatile composition and mid-infrared spectroscopic profiles of South African young cultivar wines with chemometrics to identify compositional trends and to distinguish between the different cultivars. Data were generated by gas chromatography and FTMIR spectroscopy and investigated by using analysis of variance (ANOVA), principal component analysis (PCA), and linear discriminant analysis (LDA). Significant differences were found in the volatile composition of the cultivar wines, with marked similarities in the composition of Pinotage wines and white wines, specifically for 2-phenylethanol, butyric acid, ethyl acetate, isoamyl acetate, isoamyl alcohol, and isobutyric acid. Of the 26 compounds that were analyzed, 14 had odor activity values of >1. The volatile composition and FTMIR spectra both contributed to the differentiation between the cultivar wines. The best discrimination model between the white wines was based on FTMIR spectra (98.3% correct classification), whereas a combination of spectra and volatile compounds (86.8% correct classification) was best to discriminate between the red wine cultivars.