Qualitative Evaluation of Common Quantitative Metrics for Clinical Acceptance of Automatic Segmentation: a Case Study on Heart Contouring from CT Images by Deep Learning Algorithms.

Organs-at-risk contouring is time consuming and labour intensive. Automation by deep learning algorithms would decrease the workload of radiotherapists and technicians considerably. However, the variety of metrics used for the evaluation of deep learning algorithms make the results of many papers difficult to interpret and compare. In this paper, a qualitative evaluation is done on five established metrics to assess whether their values correlate with clinical usability. A total of 377 CT volumes with heart delineations were randomly selected for training and evaluation. A deep learning algorithm was used to predict the contours of the heart. A total of 101 CT slices from the validation set with the predicted contours were shown to three experienced radiologists. They examined each slice independently whether they would accept or adjust the prediction and if there were (small) mistakes. For each slice, the scores of this qualitative evaluation were then compared with the Sørensen-Dice coefficient (DC), the Hausdorff distance (HD), pixel-wise accuracy, sensitivity and precision. The statistical analysis of the qualitative evaluation and metrics showed a significant correlation. Of the slices with a DC over 0.96 (N = 20) or a 95% HD under 5 voxels (N = 25), no slices were rejected by the readers. Contours with lower DC or higher HD were seen in both rejected and accepted contours. Qualitative evaluation shows that it is difficult to use common quantification metrics as indicator for use in clinic. We might need to change the reporting of quantitative metrics to better reflect clinical acceptance.

Automatic Cardiac Structure Contouring for Small Datasets with Cascaded Deep Learning Models.

Cardiac structure contouring is a time consuming and tedious manual activity used for radiotherapeutic dose toxicity planning. We developed an automatic cardiac structure segmentation pipeline for use in low-dose non-contrast planning CT based on deep learning algorithms for small datasets. Fifty CT scans were retrospectively selected and the whole heart, ventricles and atria were contoured. A two stage deep learning pipeline was trained on 41 non contrast planning CTs, tuned with 3 CT scans and validated on 6 CT scans. In the first stage, An InceptionResNetV2 network was used to identify the slices that contained cardiac structures. The second stage consisted of three deep learning models trained on the images containing cardiac structures to segment the structures. The three deep learning models predicted the segmentations/contours on axial, coronal and sagittal images and are combined to create the final prediction. The final accuracy of the pipeline was quantified on 6 volumes by calculating the Dice similarity coefficient (DC), 95% Hausdorff distance (95% HD) and volume ratios between predicted and ground truth volumes. Median DC and 95% HD of 0.96, 0.88, 0.92, 0.80 and 0.82, and 1.86, 2.98, 2.02, 6.16 and 6.46 were achieved for the whole heart, right and left ventricle, and right and left atria respectively. The median differences in volume were -4, -1, + 5, -16 and -20% for the whole heart, right and left ventricle, and right and left atria respectively. The automatic contouring pipeline achieves good results for whole heart and ventricles. Robust automatic contouring with deep learning methods seems viable for local centers with small datasets.

Changes in quantitative CT image features of ground-glass nodules in differentiating invasive pulmonary adenocarcinoma from benign and in situ lesions: histopathological comparisons.

AIM: To evaluate progressive changes in quantitative CT features of the non-solid component of ground-glass nodules (GGNs) from baseline to follow-up to differentiate invasive (minimally invasive adenocarcinoma [MIA] and invasive adenocarcinoma [IA]) GGNs from benign or pre-invasive (adenocarcinoma in situ [AIS]) lesions. MATERIALS AND METHODS: Patients with a GGN detected at baseline and follow-up computed tomography (CT), examined by tissue sampling were included in the study. The diameter and mean, maximum, minimum CT density and density deviation from the non-solid component of whole GGNs were measured. Progression of these features over time was analysed by linear regression analysis. Multivariate receiver operating characteristics analyses of combined measures created by a logistic regression model were performed to evaluate diagnostic performance for invasive GGNs. RESULTS: Sixty-one patients (24 males) with 70 GGNs were included. Fifteen GGNs were benign, six AIS, 38 MIA, and 11 IA. The mean diameter of all histological subtypes increased from baseline to follow-up, the largest increase was found in the MIA group (p<0.001). For MIA and IA, the mean, maximum, minimum density, and density deviation had a positive correlation over time, whilst benign and pre-invasive GGNs showed a negative correlation for these features. A diagnostic model based on three GGN features (increasing in diameter, mean density, and density deviation) identified invasive GGNs with a sensitivity, specificity and area under the receiver operating characteristic (ROC) curve (AUC) of 83.7%, 61.9%, and 0.786, respectively (p<0.001). CONCLUSION: In GGN follow-up, the diameter of benign and AIS, and invasive GGNs significantly increased. Additional analysis of mean density and density deviation in the non-solid component may help to identify invasive GGNs.

Complication rates of CT-guided transthoracic lung biopsy: meta-analysis.

OBJECTIVES: To meta-analyze complication rate in computed tomography (CT)-guided transthoracic lung biopsy and associated risk factors. METHODS: Four databases were searched from 1/2000 to 8/2015 for studies reporting complications in CT-guided lung biopsy. Overall and major complication rates were pooled and compared between core biopsy and fine needle aspiration (FNA) using the random-effects model. Risk factors for complications in core biopsy and FNA were identified in meta-regression analysis. RESULTS: For core biopsy, 32 articles (8,133 procedures) were included and for FNA, 17 (4,620 procedures). Pooled overall complication rates for core biopsy and FNA were 38.8 % (95 % CI: 34.3-43.5 %) and 24.0 % (95 % CI: 18.2-30.8 %), respectively. Major complication rates were 5.7 % (95 % CI: 4.4-7.4 %) and 4.4 % (95 % CI: 2.7-7.0 %), respectively. Overall complication rate was higher for core biopsy compared to FNA (p < 0.001). For FNA, larger needle diameter was a risk factor for overall complications, and increased traversed lung parenchyma and smaller lesion size were risk factors for major complications. For core biopsy, no significant risk factors were identified. CONCLUSIONS: In CT-guided lung biopsy, minor complications were common and occurred more often in core biopsy than FNA. Major complication rate was low. For FNA, smaller nodule diameter, larger needle diameter and increased traversed lung parenchyma were risk factors for complications. KEY POINTS: • Minor complications are common in CT-guided lung biopsy • Major complication rate is low in CT-guided lung biopsy • CT-guided lung biopsy complications occur more often in core biopsy than FNA • Major complication rate is similar in core biopsy and FNA • Risk factors for FNA are larger needle diameter, smaller lesion size.

Diagnostic performance of semi-quantitative and quantitative stress CMR perfusion analysis: a meta-analysis.

BACKGROUND: Stress cardiovascular magnetic resonance (CMR) perfusion imaging is a promising modality for the evaluation of coronary artery disease (CAD) due to high spatial resolution and absence of radiation. Semi-quantitative and quantitative analysis of CMR perfusion are based on signal-intensity curves produced during the first-pass of gadolinium contrast. Multiple semi-quantitative and quantitative parameters have been introduced. Diagnostic performance of these parameters varies extensively among studies and standardized protocols are lacking. This study aims to determine the diagnostic accuracy of semi- quantitative and quantitative CMR perfusion parameters, compared to multiple reference standards. METHOD: Pubmed, WebOfScience, and Embase were systematically searched using predefined criteria (3272 articles). A check for duplicates was performed (1967 articles). Eligibility and relevance of the articles was determined by two reviewers using pre-defined criteria. The primary data extraction was performed independently by two researchers with the use of a predefined template. Differences in extracted data were resolved by discussion between the two researchers. The quality of the included studies was assessed using the 'Quality Assessment of Diagnostic Accuracy Studies Tool' (QUADAS-2). True positives, false positives, true negatives, and false negatives were subtracted/calculated from the articles. The principal summary measures used to assess diagnostic accuracy were sensitivity, specificity, andarea under the receiver operating curve (AUC). Data was pooled according to analysis territory, reference standard and perfusion parameter. RESULTS: Twenty-two articles were eligible based on the predefined study eligibility criteria. The pooled diagnostic accuracy for segment-, territory- and patient-based analyses showed good diagnostic performance with sensitivity of 0.88, 0.82, and 0.83, specificity of 0.72, 0.83, and 0.76 and AUC of 0.90, 0.84, and 0.87, respectively. In per territory analysis our results show similar diagnostic accuracy comparing anatomical (AUC 0.86(0.83-0.89)) and functional reference standards (AUC 0.88(0.84-0.90)). Only the per territory analysis sensitivity did not show significant heterogeneity. None of the groups showed signs of publication bias. CONCLUSIONS: The clinical value of semi-quantitative and quantitative CMR perfusion analysis remains uncertain due to extensive inter-study heterogeneity and large differences in CMR perfusion acquisition protocols, reference standards, and methods of assessment of myocardial perfusion parameters. For wide spread implementation, standardization of CMR perfusion techniques is essential. TRIAL REGISTRATION: CRD42016040176 .

Image De-Identification Methods for Clinical Research in the XDS Environment.

To investigate possible de-identification methodologies within the Cross-Enterprise Document Sharing for imaging (XDS-I) environment in order to provide strengthened support for image data exchange as part of clinical research projects. De-identification, using anonymization or pseudonymization, is the most common method to perform information removal within DICOM data. However, it is not a standard part of the XDS-I profiles. Different methodologies were observed to define how and where de-identification should take place within an XDS environment used for scientific research. De-identification service can be placed in three locations within the XDS-I framework: 1) within the Document Source, 2) between the Document Source and Document Consumer, and 3) within the Document Consumer. First method has a potential advantage with respect to the exposure of the images to outside systems but has drawbacks with respect to additional hardware and configuration requirements. Second and third method have big concern in exposing original documents with all identifiable data being intact after leaving the Document Source. De-identification within the Document Source has more advantages compared to the other methods. On the contrary, it is less recommended to perform de-identification within the Document Consumer since it has the highest risk of the exposure of patients identity due to the fact that images are exposed without de-identification during the transfers.

Free DICOM de-identification tools in clinical research: functioning and safety of patient privacy.

PURPOSE: To compare non-commercial DICOM toolkits for their de-identification ability in removing a patient's personal health information (PHI) from a DICOM header. MATERIALS AND METHODS: Ten DICOM toolkits were selected for de-identification tests. Tests were performed by using the system's default de-identification profile and, subsequently, the tools' best adjusted settings. We aimed to eliminate fifty elements considered to contain identifying patient information. The tools were also examined for their respective methods of customization. RESULTS: Only one tool was able to de-identify all required elements with the default setting. Not all of the toolkits provide a customizable de-identification profile. Six tools allowed changes by selecting the provided profiles, giving input through a graphical user interface (GUI) or configuration text file, or providing the appropriate command-line arguments. Using adjusted settings, four of those six toolkits were able to perform full de-identification. CONCLUSION: Only five tools could properly de-identify the defined DICOM elements, and in four cases, only after careful customization. Therefore, free DICOM toolkits should be used with extreme care to prevent the risk of disclosing PHI, especially when using the default configuration. In case optimal security is required, one of the five toolkits is proposed. KEY POINTS: • Free DICOM toolkits should be carefully used to prevent patient identity disclosure. • Each DICOM tool produces its own specific outcomes from the de-identification process. • In case optimal security is required, using one DICOM toolkit is proposed.

Osteoporosis markers on low-dose lung cancer screening chest computed tomography scans predict all-cause mortality.

OBJECTIVES: Further survival benefits may be gained from low-dose chest computed tomography (CT) by assessing vertebral fractures and bone density. We sought to assess the association between CT-measured vertebral fractures and bone density with all-cause mortality in lung cancer screening participants. METHODS: Following a case-cohort design, lung cancer screening trial participants (N = 3,673) who died (N = 196) during a median follow-up of 6 years (inter-quartile range: 5.7-6.3) were identified and added to a random sample of N = 383 from the trial. We assessed vertebral fractures using Genant's semiquantative method on sagittal reconstructions and measured bone density (Hounsfield Units (HU)) in vertebrae. Cox proportional hazards modelling was used to determine if vertebral fractures or bone density were independently predictive of mortality. RESULTS: The prevalence of vertebral fractures was 35% (95% confidence interval 30-40%) among survivors and 51% (44-58%) amongst cases. After adjusting for age, gender, smoking status, pack years smoked, coronary and aortic calcium volume and pulmonary emphysema, the adjusted hazard ratio (HR) for vertebral fracture was 2.04 (1.43-2.92). For each 10 HU decline in trabecular bone density, the adjusted HR was 1.08 (1.02-1.15). CONCLUSIONS: Vertebral fractures and bone density are independently associated with all-cause mortality. KEY POINTS: • Lung cancer screening chest computed tomography contains additional, potentially useful information. • Vertebral fractures and bone density are independently predictive of mortality. • This finding has implications for screening and management decisions.

A web-based institutional DICOM distribution system with the integration of the Clinical Trial Processor (CTP).

To develop and test a fast and easy rule-based web-environment with optional de-identification of imaging data to facilitate data distribution within a hospital environment. A web interface was built using Hypertext Preprocessor (PHP), an open source scripting language for web development, and Java with SQL Server to handle the database. The system allows for the selection of patient data and for de-identifying these when necessary. Using the services provided by the RSNA Clinical Trial Processor (CTP), the selected images were pushed to the appropriate services using a protocol based on the module created for the associated task. Five pipelines, each performing a different task, were set up in the server. In a 75 month period, more than 2,000,000 images are transferred and de-identified in a proper manner while 20,000,000 images are moved from one node to another without de-identification. While maintaining a high level of security and stability, the proposed system is easy to setup, it integrate well with our clinical and research practice and it provides a fast and accurate vendor-neutral process of transferring, de-identifying, and storing DICOM images. Its ability to run different de-identification processes in parallel pipelines is a major advantage in both clinical and research setting.

Prevalence of severe subclinical coronary artery disease on cardiac CT and MRI in patients with extra-cardiac arterial disease.

OBJECTIVE: Patients with extra-cardiac arterial disease (ECAD) are at high risk of coronary artery disease (CAD). Prevalence of silent, significant CAD in patients with stenotic or aneurysmal ECAD was examined. Early detection and treatment may reduce CAD mortality in this high-risk group. MATERIALS AND METHODS: ECAD patients without cardiac complaints underwent computed tomography (CT) for calcium scoring, coronary CT angiography (cCTA) if calcium score was 1,000 or under, and adenosine perfusion magnetic resonance imaging (APMR) if there was no left main stenosis. Significant CAD was defined as calcium score over 1,000, cCTA-detected coronary stenosis of at least 50% lumen diameter, and/or APMR-detected inducible myocardial ischemia. In cases of left main stenosis (or equivalent) or myocardial ischemia, patients were referred to a cardiologist. RESULTS: The prevalence of significant CAD was 56.8% (95% CI 47.5 to 66.0). One-hundred and eleven patients were included. Eighty-four patients (76%) had stenotic ECAD, and 27 (24%) had aneurysmal disease. In patients with stenotic ECAD, significant coronary stenosis was present in 32 (38%) and inducible ischemia in eight (12%). Corresponding results in aneurysmal ECAD were eight (30%) and two (11%), respectively (p for difference >.05). Sixteen (19%) patients with stenotic and six (22%) with aneurysmal ECAD were referred to a cardiologist, with subsequent cardiac intervention in seven (44%) and three (50%), respectively (both p >.05). CONCLUSIONS: Patients with stenotic or aneurysmal ECAD have a high prevalence of silent, significant CAD.

Morphological characteristics of potentially malignant pulmonary nodules in high-risk male smokers detected in lung cancer screening trial in Cracow, Poland.

The purpose of this paper was to present morphological characteristics of potentially malignant nodules revealed in a group of male smokers aged 50-74 with a very high risk for developing lung cancer estimated in the study for lung cancer screening in Cracow (Poland). Nine hundred male smokers aged 50 to 74 years were invited to the study and were asked in questionnaires about e.g. smoking exposure history. Exclusion criteria included e.g. positive cancer history and chest computed tomography (CT) examination in the previous year. Based on CT results and characteristics of pulmonary nodules subjects were classified to group A (low risk), group B (indeterminate) and group C (high-risk individuals - required work-up). Final diagnosis was based on pathological results of postoperative material. Thirty-nine males of mean age 63.4 (standard deviation (SD): 6.69 years) revealed 41 potentially malignant pulmonary nodules in baseline screening. In 14 subjects 16 type C pulmonary nodules were histologically proved. Nine nodules were found to be benign lesions, while 7 nodules revealed malignant lung cancer: 5 cases of adenocarcinoma and 2 cases of adenosquamous carcinoma. We determined morphological characteristics of potentially malignant pulmonary nodules in 39 high-risk male smokers and proved lung cancer in 7 subjects.

Influenza season influence on outcome of new nodules in the NELSON study.

We evaluated the impact of the influenza season on outcome of new lung nodules in a LDCT lung cancer screening trial population. NELSON-trial participants with ≥ 1 new nodule detected in screening rounds two and three were included. Outcome (resolution or persistence) of new nodules detected per season was calculated and compared. Winter (influenza season) was defined as 1st October to 31st March, and compared to the summer (hay-fever season), 1st April to 30th September. Overall, 820 new nodules were reported in 529 participants. Of the total new nodules, 482 (59%) were reported during winter. When considering the outcome of all new nodules, there was no statistically significant association between summer and resolving nodules (OR 1.07 [CI 1.00-1.15], p = 0.066), also when looking at the largest nodule per participant (OR 1.37 [CI 0.95-1.98], p = 0.094). Similarly, there was no statistically significant association between season and screen detected cancers (OR 0.47 [CI 0.18-1.23], p = 0.123). To conclude, in this lung cancer screening population, there was no statistically significant association between influenza season and outcome of new lung nodules. Hence, we recommend new nodule management strategy is not influenced by the season in which the nodule is detected.

The added value of quantitative multi-voxel MR spectroscopy in breast magnetic resonance imaging.

OBJECTIVE: To determine whether quantitative multivoxel MRS improves the accuracy of MRI in the assessment of breast lesions. METHODS: Twenty-five consecutive patients with 26 breast lesions ≥ 1 cm assessed as BI-RADS 3 or 4 with mammography underwent quantitative multivoxel MRS and contrast-enhanced MRI. The choline (Cho) concentration was calculated using the unsuppressed water signal as a concentration reference. ROC analysis established the diagnostic accuracy of MRI and MRS in the assessment of breast lesions. RESULTS: Respective Cho concentrations in 26 breast lesions re-classified by MRI as BI-RADS 2 (n = 5), 3 (n = 8), 4 (n = 5) and 5 (n = 8) were 1.16 ± 0.43 (mean ± SD), 1.43 ± 0.47, 2.98 ± 2.15 and 4.94 ± 3.10 mM. Two BI-RADS 3 lesions and all BI-RADS 4 and 5 lesions were malignant on histopathology and had Cho concentrations between 1.7 and 11.8 mM (4.03 ± 2.72 SD), which were significantly higher (P = 0.01) than that in the 11 benign lesions (0.4-1.5 mM; 1.19 ± 0.33 SD). Furthermore, Cho concentrations in the benign and malignant breast lesions in BI-RADS 3 category differed (P = 0.01). The accuracy of combined multivoxel MRS/breast MRI BI-RADS re-classification (AUC = 1.00) exceeded that of MRI alone (AUC = 0.96 ± 0.03). CONCLUSIONS: These preliminary data indicate that multivoxel MRS improves the accuracy of MRI when using a Cho concentration cut-off ≤ 1.5 mM for benign lesions. KEY POINTS: Quantitative multivoxel MR spectroscopy can improve the accuracy of contrast-enhanced breast MRI. Multivoxel-MRS can differentiate breast lesions by using the highest Cho-concentration. Multivoxel-MRS can exclude patients with benign breast lesions from further invasive diagnostic procedures.

MR intensity measurements of nondenervated muscle in patients following severe forearm trauma.

Fluid increases resulting in higher MRI signal intensities in T(2) -weighted and short tau inversion recovery (STIR) sequences can be used to diagnose nerve injury. By comparing the signal intensities over time, MRI may become a new method for monitoring the healing process. Muscle edema is assessed by comparing the signal intensity of affected muscle with that of nonaffected muscle. However, in severe forearm trauma, the signal of nondenervated muscle may also be increased by wound edema, thus masking the effect of denervation. Hence, the purpose of this study was to investigate the influence of wound edema on muscle signal intensity in 29 consecutive patients examined on a 1.5-T MRI scanner at 1, 3, 6, 9 and 12 months after severe forearm trauma. The long-term course of wound edema and the influence of wound distance were thus investigated using a standardized imaging, calibration and post-processing protocol. The signal intensities of nondenervated intrinsic hand muscles were measured in the affected and contralateral sides. Muscle signal intensities were increased on the trauma side at 1 and 3 months (18% and 7.4%, respectively; p < 0.001) and normalized thereafter. In the contralateral hand, no significant signal changes were seen. No relationship was found between wound distance and the severity of wound edema. This study shows that wound edema influences muscle signal intensity comparisons in patients with forearm trauma. When comparing denervated muscle with nondenervated muscle, an additional scan of the contralateral side is indicated during the first 6 months after trauma to assess the extent of wound edema. After 6 months, the ipsilateral side can be used for muscle signal intensity comparisons.

Feasibility of computed tomography based thermometry during interstitial laser heating in bovine liver.

OBJECTIVES: To assess the feasibility of computed tomography (CT) based thermometry during interstitial laser heating in the bovine liver. METHODS: Four freshly exercised cylindrical blocks of bovine tissue were heated using a continuous laser of Nd:YAG (wavelength: 1064 nm, active length: 30 mm, power: 10-30 W). All tissues were imaged at least once before and 7 times during laser heating using CT and temperatures were simultaneously measured with 5 calibrated thermal sensors. The dependency of the average CT numbers as a function of temperature was analysed with regression analysis and a CT thermal sensitivity was derived. RESULTS: During laser heating, the growing hypodense area was observed around the laser source and that area showed an increase as a function of time. The formation of hypodense area was caused by declining in CT numbers at increasing temperatures. The regression analysis showed an inverse linear dependency between temperature and average CT number with -0.65 ± 0.048 HU/°C (R(2) = 0.75) for the range of 18-85°C in bovine liver. CONCLUSIONS: The non-invasive CT based thermometry during interstitial laser heating is feasible in the bovine liver. CT based thermometry could be further developed and may be of potential use during clinical LITT of the liver.

Diffusion weighted imaging of the breast: Performance of standardized breast tumor tissue selection methods in clinical decision making.

OBJECTIVES: In breast diffusion weighted imaging (DWI) protocol standardization, it is recently shown that no breast tumor tissue selection (BTTS) method outperformed the others. The purpose of this study is to analyze the feasibility of three fixed-size breast tumor tissue selection (BTTS) methods based on the reproducibility, accuracy and time-measurement in comparison to the largest oval and manual delineation in breast diffusion weighted imaging data. METHODS: This study is performed with a consecutive dataset of 116 breast lesions (98 malignant) of at least 1.0 cm, scanned in accordance with the EUSOBI breast DWI working group recommendations. Reproducibility of the maximum size manual (BTTS1) and of the maximal size round/oval (BTTS2) methods were compared with three smaller fixed-size circular BTTS methods in the middle of each lesion (BTTS3, 0.12 cm3 volume) and at lowest apparent diffusion coefficient (ADC) (BTTS4, 0.12 cm3; BTTS5, 0.24 cm3). Mean ADC values, intraclass-correlation-coefficients (ICCs), area under the curve (AUC) and measurement times (sec) of the 5 BTTS methods were assessed by two observers. RESULTS: Excellent inter- and intra-observer agreement was found for any BTTS (with ICC 0.88-0.92 and 0.92-0.94, respectively). Significant difference in ADCmean between any pair of BTTS methods was shown (p = <0.001-0.009), except for BTTS2 vs. BTTS3 for observer 1 (p = 0.10). AUCs were comparable between BTTS methods, with highest AUC for BTTS2 (0.89-0.91) and lowest for BTTS4 (0.76-0.85). However, as an indicator of clinical feasibility, BTTS2-3 showed shortest measurement times (10-15 sec) compared to BTTS1, 4-5 (19-39 sec). CONCLUSION: The performance of fixed-size BTTS methods, as a potential tool for clinical decision making, shows equal AUC but shorter ADC measurement time compared to manual or oval whole lesion measurements. The advantage of a fixed size BTTS method is the excellent reproducibility. A central fixed breast tumor tissue volume of 0.12 cm3 is the most feasible method for use in clinical practice.

Diagnostic performance of breast tumor tissue selection in diffusion weighted imaging: A systematic review and meta-analysis.

BACKGROUND: Several methods for tumor delineation are used in literature on breast diffusion weighted imaging (DWI) to measure the apparent diffusion coefficient (ADC). However, in the process of reaching consensus on breast DWI scanning protocol, image analysis and interpretation, still no standardized optimal breast tumor tissue selection (BTTS) method exists. Therefore, the purpose of this study is to assess the impact of BTTS methods on ADC in the discrimination of benign from malignant breast lesions in DWI in terms of sensitivity, specificity and area under the curve (AUC). METHODS AND FINDINGS: In this systematic review and meta-analysis, adhering to the PRISMA statement, 61 studies, with 65 study subsets, in females with benign or malignant primary breast lesions (6291 lesions) were assessed. Studies on DWI, quantified by ADC, scanned on 1.5 and 3.0 Tesla and using b-values 0/50 and ≥ 800 s/mm2 were included. PubMed and EMBASE were searched for studies up to 23-10-2019 (n = 2897). Data were pooled based on four BTTS methods (by definition of measured region of interest, ROI): BTTS1: whole breast tumor tissue selection, BTTS2: subtracted whole breast tumor tissue selection, BTTS3: circular breast tumor tissue selection and BTTS4: lowest diffusion breast tumor tissue selection. BTTS methods 2 and 3 excluded necrotic, cystic and hemorrhagic areas. Pooled sensitivity, specificity and AUC of the BTTS methods were calculated. Heterogeneity was explored using the inconsistency index (I2) and considering covariables: field strength, lowest b-value, image of BTTS selection, pre-or post-contrast DWI, slice thickness and ADC threshold. Pooled sensitivity, specificity and AUC were: 0.82 (0.72-0.89), 0.79 (0.65-0.89), 0.88 (0.85-0.90) for BTTS1; 0.91 (0.89-0.93), 0.84 (0.80-0.87), 0.94 (0.91-0.96) for BTTS2; 0.89 (0.86-0.92), 0.90 (0.85-0.93), 0.95 (0.93-0.96) for BTTS3 and 0.90 (0.86-0.93), 0.84 (0.81-0.87), 0.86 (0.82-0.88) for BTTS4, respectively. Significant heterogeneity was found between studies (I2 = 95). CONCLUSIONS: None of the breast tissue selection (BTTS) methodologies outperformed in differentiating benign from malignant breast lesions. The high heterogeneity of ADC data acquisition demands further standardization, such as DWI acquisition parameters and tumor tissue selection to substantially increase the reliability of DWI of the breast.

MR spectroscopy and diffusion tensor imaging of the brain in Sjögren-Larsson syndrome.

Diffusion tensor imaging (DTI) is reported for the first time in a patient with Sjögren-Larsson syndrome, an autosomal recessive neurocutaneous disorder. Magnetic resonance spectroscopy (MRS) revealed normal levels of choline, creatine and N-acetyl aspartate (NAA) and the characteristic lipid signals in the white matter brain tissue. Conventional MRI showed increased signal intensity around the lateral ventricles indicating abnormal myelination. DTI revealed normal apparent diffusion coefficient (ADC) values, but reduced fractional anisotropy (FA) in the white matter. After co-registration of the parameters obtained with DTI with the results of MRS (36 voxels), significant correlations were obtained of lipid content with FA (r=0.81), ADC (r=-0.62), choline (r=0.51), and NAA (r=0.44) (P<0.01, all). These results suggest that in Sjögren-Larsson syndrome, the white matter lipid signals originate from the neurons, with NAA and choline reflecting neuron density and myelination. The comparatively high FA/low ADC values in these lipid-rich locations, indicate a loss of diffusion in directions perpendicular to the fibers. The overall loss of FA in the white matter may reflect a loss of brain tissue water content in SLS patients compared with controls and precede the formation of atrophy.