-
NeuroImage. Clinical 2021Focal cortical dysplasias (FCDs) are a common cause of apparently non-lesional drug-resistant focal epilepsy. Visual detection of subtle FCDs on MRI is clinically...
OBJECTIVE
Focal cortical dysplasias (FCDs) are a common cause of apparently non-lesional drug-resistant focal epilepsy. Visual detection of subtle FCDs on MRI is clinically important and often challenging. In this study, we implement a set of 3D local image filters adapted from computer vision applications to characterize the appearance of normal cortex surrounding the gray-white junction. We create a normative model to serve as the basis for a novel multivariate constrained outlier approach to automated FCD detection.
METHODS
Standardized MPRAGE, T and FLAIR MR images were obtained in 15 patients with radiologically or histologically diagnosed FCDs and 30 healthy volunteers. Multiscale 3D local image filters were computed for each MR contrast then sampled onto the gray-white junction surface. Using an iterative Gaussianization procedure, we created a normative model of cortical variability in healthy volunteers, allowing for identification of outlier regions and estimates of similarity in normal cortex and FCD lesions. We used a constrained outlier approach following local normalization to automatically detect FCD lesions based on projection onto the mean FCD feature vector.
RESULTS
FCDs as well as some normal cortical regions such as primary sensorimotor and paralimbic regions appear as outliers. Regions such as the paralimbic regions and the anterior insula have similar features to FCDs. Our constrained outlier approach allows for automated FCD detection with 80% sensitivity and 70% specificity.
SIGNIFICANCE
A normative model using multiscale local image filters can be used to describe the normal cortical variability. Although FCDs appear similar to some cortical regions such as the anterior insula and paralimbic cortices, they can be identified using a constrained outlier detection approach. Our method for detecting outliers and estimating similarity is generic and could be extended to identification of other types of lesions or atypical cortical areas.
Topics: Epilepsy; Humans; Imaging, Three-Dimensional; Magnetic Resonance Imaging; Malformations of Cortical Development; Malformations of Cortical Development, Group I
PubMed: 33556791
DOI: 10.1016/j.nicl.2021.102565 -
NeuroImage Dec 2021The ratio of T1-weighted (T1w) and T2-weighted (T2w) magnetic resonance imaging (MRI) images is often used as a proxy measure of cortical myelin. However, the... (Comparative Study)
Comparative Study
BACKGROUND
The ratio of T1-weighted (T1w) and T2-weighted (T2w) magnetic resonance imaging (MRI) images is often used as a proxy measure of cortical myelin. However, the T1w/T2w-ratio is based on signal intensities that are inherently non-quantitative and known to be affected by extrinsic factors. To account for this a variety of processing methods have been proposed, but a systematic evaluation of their efficacy is lacking. Given the dependence of the T1w/T2w-ratio on scanner hardware and T1w and T2w protocols, it is important to ensure that processing pipelines perform well also across different sites.
METHODS
We assessed a variety of processing methods for computing cortical T1w/T2w-ratio maps, including correction methods for nonlinear field inhomogeneities, local outliers, and partial volume effects as well as intensity normalisation. These were implemented in 33 processing pipelines which were applied to four test-retest datasets, with a total of 170 pairs of T1w and T2w images acquired on four different MRI scanners. We assessed processing pipelines across datasets in terms of their reproducibility of expected regional distributions of cortical myelin, lateral intensity biases, and test-retest reliability regionally and across the cortex. Regional distributions were compared both qualitatively with histology and quantitatively with two reference datasets, YA-BC and YA-B1+, from the Human Connectome Project.
RESULTS
Reproducibility of raw T1w/T2w-ratio distributions was overall high with the exception of one dataset. For this dataset, Spearman rank correlations increased from 0.27 to 0.70 after N3 bias correction relative to the YA-BC reference and from -0.04 to 0.66 after N4ITK bias correction relative to the YA-B1+ reference. Partial volume and outlier corrections had only marginal effects on the reproducibility of T1w/T2w-ratio maps and test-retest reliability. Before intensity normalisation, we found large coefficients of variation (CVs) and low intraclass correlation coefficients (ICCs), with total whole-cortex CV of 10.13% and whole-cortex ICC of 0.58 for the raw T1w/T2w-ratio. Intensity normalisation with WhiteStripe, RAVEL, and Z-Score improved total whole-cortex CVs to 5.91%, 5.68%, and 5.19% respectively, whereas Z-Score and Least Squares improved whole-cortex ICCs to 0.96 and 0.97 respectively.
CONCLUSIONS
In the presence of large intensity nonuniformities, bias field correction is necessary to achieve acceptable correspondence with known distributions of cortical myelin, but it can be detrimental in datasets with less intensity inhomogeneity. Intensity normalisation can improve test-retest reliability and inter-subject comparability. However, both bias field correction and intensity normalisation methods vary greatly in their efficacy and may affect the interpretation of results. The choice of T1w/T2w-ratio processing method must therefore be informed by both scanner and acquisition protocol as well as the given study objective. Our results highlight limitations of the T1w/T2w-ratio, but also suggest concrete ways to enhance its usefulness in future studies.
Topics: Adult; Connectome; Datasets as Topic; Female; Humans; Image Processing, Computer-Assisted; Magnetic Resonance Imaging; Male; Middle Aged; Reproducibility of Results
PubMed: 34848300
DOI: 10.1016/j.neuroimage.2021.118709 -
Radiation Oncology (London, England) Oct 2022To save time and have more consistent contours, fully automatic segmentation of targets and organs at risk (OAR) is a valuable asset in radiotherapy. Though current deep...
AIMS
To save time and have more consistent contours, fully automatic segmentation of targets and organs at risk (OAR) is a valuable asset in radiotherapy. Though current deep learning (DL) based models are on par with manual contouring, they are not perfect and typical errors, as false positives, occur frequently and unpredictably. While it is possible to solve this for OARs, it is far from straightforward for target structures. In order to tackle this problem, in this study, we analyzed the occurrence and the possible dose effects of automated delineation outliers.
METHODS
First, a set of controlled experiments on synthetically generated outliers on the CT of a glioblastoma (GBM) patient was performed. We analyzed the dosimetric impact on outliers with different location, shape, absolute size and relative size to the main target, resulting in 61 simulated scenarios. Second, multiple segmentation models where trained on a U-Net network based on 80 training sets consisting of GBM cases with annotated gross tumor volume (GTV) and edema structures. On 20 test cases, 5 different trained models and a majority voting method were used to predict the GTV and edema. The amount of outliers on the predictions were determined, as well as their size and distance from the actual target.
RESULTS
We found that plans containing outliers result in an increased dose to healthy brain tissue. The extent of the dose effect is dependent on the relative size, location and the distance to the main targets and involved OARs. Generally, the larger the absolute outlier volume and the distance to the target the higher the potential dose effect. For 120 predicted GTV and edema structures, we found 1887 outliers. After construction of the planning treatment volume (PTV), 137 outliers remained with a mean distance to the target of 38.5 ± 5.0 mm and a mean size of 1010.8 ± 95.6 mm. We also found that majority voting of DL results is capable to reduce outliers.
CONCLUSIONS
This study shows that there is a severe risk of false positive outliers in current DL predictions of target structures. Additionally, these errors will have an evident detrimental impact on the dose and therefore could affect treatment outcome.
Topics: Humans; Radiotherapy Planning, Computer-Assisted; Glioblastoma; Organs at Risk; Radiotherapy Dosage; Radiotherapy, Intensity-Modulated
PubMed: 36273161
DOI: 10.1186/s13014-022-02137-9 -
... IEEE International Conference on... Dec 2022Outlier detection is a fundamental data analytics technique often used for many security applications. Numerous outlier detection techniques exist, and in most cases are...
Outlier detection is a fundamental data analytics technique often used for many security applications. Numerous outlier detection techniques exist, and in most cases are used to directly identify outliers without any interaction. Typically the underlying data used is often high dimensional and complex. Even though outliers may be identified, since humans can easily grasp low dimensional spaces, it is difficult for a security expert to understand/visualize why a particular event or record has been identified as an outlier. In this paper we study the extent to which outlier detection techniques work in smaller dimensions and how well dimensional reduction techniques still enable accurate detection of outliers. This can help us to understand the extent to which data can be visualized while still retaining the intrinsic outlyingness of the outliers.
PubMed: 38094985
DOI: 10.1109/tps-isa56441.2022.00028 -
Turkish Journal of Orthodontics Dec 2022To determine whether multiple siblings resemble one another in their craniofacial characteristics as measured on cephalometric radiographs.
OBJECTIVE
To determine whether multiple siblings resemble one another in their craniofacial characteristics as measured on cephalometric radiographs.
METHODS
This study was conducted retrospectively using the Forsyth Moorrees twin sample. A total of 32 families were included, each with ≥4 postpubertal siblings, totaling 142 subjects. Only 1 monozygotic twin was included per family. Headfilms were digitized, skeletal landmarks were located, and 6 parameters that indicated sagittal jaw relationships and vertical status were measured. Diverse statistical approaches were used. Dixon's Q-test detected outliers in a family for a given parameter. Manhattan Distance quantified similarity among siblings per parameter. Scatter plots visually displayed subject's measure relative to the mean and standard deviation of each parameter to assess the clinical relevance of the differences.
RESULTS
A total of 11 families (34.4%) had no outliers on any parameter, 13 families (40.6%) had outliers on 1 parameter, and 8 families (25%) had outliers on ≥2 parameters. We identified 29 individuals with at least 1 outlying measure (20.4%). Among these, only 2 individuals (1.4%) were significantly different from their siblings for more than 1 measurement. Although the majority of the families did not demonstrate any statistical outlier, the ranges of the measurements were clinically relevant as they might suggest different treatment. For example, the mean range of SNB (Sella-Nasion-B point) angles was 7.23°, and the mean range of MPA was 9.42°.
CONCLUSION
Although families are generally not dissimilar in their craniofacial characteristics, measurements from siblings cannot be used to predict the measurements of another sibling in a clinically meaningful way.
PubMed: 36594544
DOI: 10.5152/TurkJOrthod.2022.21237 -
Annals of the New York Academy of... Sep 2020Convergent evolution, where independent lineages evolve similar phenotypes in response to similar challenges, can provide valuable insight into how selection operates... (Review)
Review
Convergent evolution, where independent lineages evolve similar phenotypes in response to similar challenges, can provide valuable insight into how selection operates and the limitations it encounters. However, it has only recently become possible to explore how convergent evolution is reflected at the genomic level. The overlapping outlier approach (OOA), where genome scans of multiple independent lineages are used to find outliers that overlap and therefore identify convergently evolving loci, is becoming popular. Here, we present a quantitative analysis of 34 studies that used this approach across many sampling designs, taxa, and sampling intensities. We found that OOA studies with increased biological sampling power within replicates have increased likelihood of finding overlapping, "convergent" signals of adaptation between them. When identifying convergent loci as overlapping outliers, it is tempting to assume that any false-positive outliers derived from individual scans will fail to overlap across replicates, but this cannot be guaranteed. We highlight how population demographics and genomic context can contribute toward both true convergence and false positives in OOA studies. We finish with an exploration of emerging methods that couple genome scans with phenotype and environmental measures, leveraging added information from genome data to more directly test hypotheses of the likelihood of convergent evolution.
Topics: Acclimatization; Adaptation, Physiological; Animals; Animals, Wild; Biological Evolution; Evolution, Molecular; Genome; Genomics
PubMed: 31241191
DOI: 10.1111/nyas.14177 -
JCO Clinical Cancer Informatics Oct 2022Artificial intelligence (AI) models for medical image diagnosis are often trained and validated on curated data. However, in a clinical setting, images that are outliers...
PURPOSE
Artificial intelligence (AI) models for medical image diagnosis are often trained and validated on curated data. However, in a clinical setting, images that are outliers with respect to the training data, such as those representing rare disease conditions or acquired using a slightly different setup, can lead to wrong decisions. It is not practical to expect clinicians to be trained to discount results for such outlier images. Toward clinical deployment, we have designed a method to train cautious AI that can automatically flag outlier cases.
MATERIALS AND METHODS
Our method-ClassClust-forms tight clusters of training images using supervised contrastive learning, which helps it identify outliers during testing. We compared ClassClust's ability to detect outliers with three competing methods on four publicly available data sets covering pathology, dermatoscopy, and radiology. We held out certain diseases, artifacts, and types of images from training data and examined the ability of various models to detect these as outliers during testing. We compared the decision accuracy of the models on held-out nonoutlier images also. We visualized the regions of the images that the models used for their decisions.
RESULTS
Area under receiver operating characteristic curve for outlier detection was consistently higher using ClassClust compared with the previous methods. Average accuracy on held-out nonoutlier images was also higher, and the visualizations of image regions were more informative using ClassClust.
CONCLUSION
The ability to flag outlier test cases need not be at odds with the ability to accurately classify nonoutliers in AI models. Although the latter capability has received research and regulatory attention, AI models for clinical deployment should possess the former as well.
Topics: Artificial Intelligence; Data Collection; Humans; ROC Curve; Trust
PubMed: 36228179
DOI: 10.1200/CCI.22.00067 -
Knee Surgery, Sports Traumatology,... Feb 2022In total knee arthroplasty (TKA), implants are increasingly aligned based on emerging patient-specific alignment strategies, such as unrestricted kinematic alignment... (Review)
Review
PURPOSE
In total knee arthroplasty (TKA), implants are increasingly aligned based on emerging patient-specific alignment strategies, such as unrestricted kinematic alignment (KA), according to their constitutional limb alignment (phenotype alignment), which results in a large proportion of patients having a hip-knee angle (HKA) outside the safe range of ± 3° to 180° traditionally considered in the mechanical alignment strategy. The aim of this systematic review is to investigate whether alignment outside the safe zone of ± 3° is associated with a higher revision rate and worse clinical outcome than alignment within this range.
METHODS
A systematic literature search was conducted in PubMed, Embase, Cochrane and World of Science, with search terms including synonyms and plurals for "total knee arthroplasty", "alignment", "outlier", "malalignment", "implant survival" and "outcome". Five studies were identified with a total number of 927 patients and 952 implants. The Oxford Knee Score (OKS) and the WOMAC were used to evaluate the clinical outcome. The follow-up period was between 6 months and 10 years.
RESULTS
According to HKA 533 knees were aligned within ± 3°, 47 (8.8%) were varus outliers and 121 (22.7%) were valgus outliers. No significant differences in clinical outcomes were found between implants positioned within ± 3° and varus and valgus outliers. Likewise, no significant differences were found regarding revision rates and implant survival.
CONCLUSION
The universal use of the "safe zone" of ± 3° derived from the mechanical alignment strategy is hardly applicable to modern personalised alignment strategies in the light of current literature. However, given the conflicting evidence in the literature on the risks of higher revision rates and poorer clinical outcomes especially with greater tibial component deviation, the lack of data on the outcomes of more extreme alignments, and regarding the use of implants for KA TKA that are actually designed for mechanical alignment, there is an urgent need for research to define eventual evidence-based thresholds for new patient-specific alignment strategies, not only for HKA but also for FMA and TMA, also taking into account the preoperative phenotype and implant design. It is of utmost clinical relevance for the application of modern alignment strategies to know which native phenotypes may be reproduced with a TKA.
LEVEL OF EVIDENCE
IV.
Topics: Arthroplasty, Replacement, Knee; Biomechanical Phenomena; Humans; Knee Joint; Knee Prosthesis; Osteoarthritis, Knee; Retrospective Studies
PubMed: 34973095
DOI: 10.1007/s00167-021-06811-5 -
JAMA Otolaryngology-- Head & Neck... Jul 2021Guidelines for many head and neck cancers, especially laryngeal cancers, allow for multiple treatment options. Currently, inequitable provision of surgery may contribute...
IMPORTANCE
Guidelines for many head and neck cancers, especially laryngeal cancers, allow for multiple treatment options. Currently, inequitable provision of surgery may contribute to outcome disparities. However, the role of geospatial factors remains understudied.
OBJECTIVE
To assess the association between US geospatial factors and treatment selection for patients with laryngeal cancer.
DESIGN, SETTING, AND PARTICIPANTS
In this retrospective cohort study, patients diagnosed with laryngeal squamous cell carcinoma between January 1, 2004, and December 31, 2014, were identified from the Surveillance, Epidemiology, and End Results database. Adjusted odds ratios (aORs) for surgical treatment were generated from multivariable, hierarchical models to assess associations with oncologic, demographic, and county variables. Outlier US counties with the highest and lowest aORs were described. Data analysis was performed from April 29 to September 11, 2020.
EXPOSURES
County of residence.
MAIN OUTCOMES AND MEASURES
The aORs for surgical treatment were generated from multivariable, hierarchical models. Outlier counties with the highest and lowest aORs are described.
RESULTS
The cohort includes 21 289 patients (mean [SD] age, 63.6 [11.2] years; 17 214 [80.9%] male) in 598 counties. Most counties had no otolaryngologist (365 [61.0%]) or radiation oncologist (434 [72.6%]). Surgery rates varied from 7.1% to 85.7% among counties with at least 10 cases. After oncologic variables were controlled for, factors independently associated with surgical treatment included patient age (aOR [95% CI], 0.94; 0.91-0.98 per 10 years), marital status (single versus married: aOR [95% CI], 0.87 [0.79-0.97]), and county social deprivation index (aOR [95% CI], 0.98 [0.97-1.00 per 5 points]) but not physician number (≥2 otolaryngologists: aOR [95% CI], 0.91 [0.75-1.11] vs ≥1 radiation oncologist: aOR [95% CI], 0.91; 0.75-1.11). The 5% of counties most likely to provide surgery (aOR, >1.23) were nearly all large metropolitan areas (2593 patients [93.3%]) and treated a disproportionately large number of patients (2778 [13.1%]). The 5% of counties least likely to provide surgery (aOR, <0.79) were also mostly large metropolitan areas (1676 patients [91.2%]) and treated a disproportionately large number of patients (1838 [8.6%]). Patients in counties least likely to provide surgery had inferior survival compared with those most likely to provide surgery (adjusted hazard ratio, 1.16; 95% CI, 1.00-1.35).
CONCLUSIONS AND RELEVANCE
These findings suggest that sociodemographic factors contribute to the wide variety in surgical treatment practices by county. The largest metropolitan counties were often outliers regarding their adjusted odds of surgical treatment. This finding is concerning for the counties least likely to provide surgery where survival is inferior.
Topics: Adult; Aged; Carcinoma, Squamous Cell; Decision Making; Demography; Female; Humans; Laryngeal Neoplasms; Male; Middle Aged; Patient Selection; Residence Characteristics; Retrospective Studies; SEER Program; United States
PubMed: 33885716
DOI: 10.1001/jamaoto.2021.0453 -
Scientific Reports Nov 2023Adverse pregnancy outcomes, such as low birth weight (LBW) and preterm birth (PTB), can have serious consequences for both the mother and infant. Early prediction of...
Adverse pregnancy outcomes, such as low birth weight (LBW) and preterm birth (PTB), can have serious consequences for both the mother and infant. Early prediction of such outcomes is important for their prevention. Previous studies using traditional machine learning (ML) models for predicting PTB and LBW have encountered two important limitations: extreme class imbalance in medical datasets and the inability to account for complex relational structures between entities. To address these limitations, we propose a node embedding-based graph outlier detection algorithm to predict adverse pregnancy outcomes. We developed a knowledge graph using a well-curated representative dataset of the Emirati population and two node embedding algorithms. The graph autoencoder (GAE) was trained by applying a combination of original risk factors and node embedding features. Samples that were difficult to reconstruct at the output of GAE were identified as outliers considered representing PTB and LBW samples. Our experiments using LBW, PTB, and very PTB datasets demonstrated that incorporating node embedding considerably improved performance, achieving a 12% higher AUC-ROC compared to traditional GAE. Our study demonstrates the effectiveness of node embedding and graph outlier detection in improving the prediction performance of adverse pregnancy outcomes in well-curated population datasets.
Topics: Pregnancy; Female; Infant, Newborn; Humans; Pregnancy Outcome; Premature Birth; Infant, Low Birth Weight; Mothers; Risk Factors
PubMed: 37963898
DOI: 10.1038/s41598-023-46726-4