Search

Automated image analysis plays an increasing role in radiology in detecting and quantifying image features outside of the perception of human eyes. Common AI-based approaches address a single medical problem, although patients often present with multiple interacting, frequently subclinical medical conditions. A holistic imaging diagnostics tool based on artificial intelligence (AI) has the potential of providing an overview of multi-system comorbidities within a single workflow. An interdisciplinary, multicentric team of medical experts and computer scientists designed a pipeline, comprising AI-based tools for the automated detection, quantification and characterization of the most common pulmonary, metabolic, cardiovascular and musculoskeletal comorbidities in chest computed tomography (CT). To provide a comprehensive evaluation of each patient, a multidimensional workflow was established with algorithms operating synchronously on a decentralized Joined Imaging Platform (JIP). The results of each patient are transferred to a dedicated database and summarized as a structured report with reference to available reference values and annotated sample images of detected pathologies. Hence, this tool allows for the comprehensive, large-scale analysis of imaging-biomarkers of comorbidities in chest CT, first in science and then in clinical routine. Moreover, this tool accommodates the quantitative analysis and classification of each pathology, providing integral diagnostic and prognostic value, and subsequently leading to improved preventive patient care and further possibilities for future studies.

Zusammenfassung Hintergrund Seit 2017 berichten Wissenschafts- und Populärmedien immer wieder von bildanalytischen Verfahren der künstlichen Intelligenz, die in der medizinischen Diagnostik zu menschlichen Experten vergleichbar gute Resultate erzielen. Mit der erstmaligen Zulassung eines solchen Systems durch die amerikanische Arzneimittelbehörde 2018 begann ihr Einzug in den klinischen Alltag. Fragestellung Dieser Beitrag gibt einen Überblick der wichtigsten Entwicklungen der künstlichen Intelligenz für bildanalytische Verfahren in klinischen Anwendungen mit Fokus auf die Dermatologie. Material und Methode Am Beispiel der ImageNet Challenge wird gezeigt, dass klassische Ansätze des maschinellen Lernens stark auf menschlicher Expertise beruhten und dass ihre Performance und Skalierbarkeit praktischen Anforderungen nicht genügen konnten. Mittels Deep Learning, einem auf neuronalen Netzen beruhenden Verfahren, konnten diese Limitierungen und insbesondere die Abhängigkeit von menschlicher Expertise überwunden werden. Wir beschreiben wichtige Eigenschaften von Deep Learning, den methodischen Durchbruch von Transfer Learning und berichten über vielversprechende Entwicklungen hin zu generativen Modellen. Ergebnisse Mittels Deep Learning erreichen bildanalytische Verfahren in vielen Fällen die für den industriellen und klinischen Einsatz geforderte Genauigkeit. Zudem gestaltet sich ihre Industrialisierung weitestgehend barrierefrei. Derzeitige Entwicklungen fokussieren sich daher weniger auf die nochmalige Verbesserung der Genauigkeit, sondern nehmen sich den Herausforderungen der Interpretierbarkeit und Anwendbarkeit unter Alltagsbedingungen an. Durch die Weiterentwicklung zu generativen Modellen werden gänzlich neuartige Anwendungen möglich. Schlussfolgerungen Deep Learning hat in vielerlei Hinsicht beeindruckende Erfolge vorzuweisen und gilt heute nicht nur in der Bildanalyse als das Standardverfahren schlechthin. Dieser Durchbruch der künstlichen Intelligenz ermöglicht eine rasch anwachsende Zahl von klinischen Anwendungen und entwickelt sich fortwährend zu einem unverzichtbaren Werkzeug in der modernen Medizin.

Picrosirius red (PSR) staining is generally used to evaluate liver fibrosis; however, PSR sometimes causes nonspecific nuclear staining. In this study, we evaluated the ability of phosphomolybdic acid (PMA) pretreatment to prevent nonspecific nuclear staining by PSR. In a manual evaluation of 27 non-tumor samples from patients with hepatocellular carcinoma, nonspecific nuclear staining was observed in 3.7% of PMA-treated specimens, compared with 85.2% of untreated specimens. Conversely, computer-assisted image analysis (CAIA) identified nonspecific nuclear staining in 0% of PMA-treated samples, vs 44.4% of untreated samples. Surprisingly, after mounting, PMA-treated specimens exhibited a blue tinge because of molybdenum blue (MB) production following sunlight exposure or virtual slide scanning. Using UV cut film, MB production induced by sunlight exposure was prevented; however, the film did not prevent MB production during virtual slide scanning. Moreover, only blue light–emitting diode exposure resulted in a blue tinge in PMA solution. Our data indicated that PMA pretreatment is effective for evaluating liver fibrosis using CAIA. Meanwhile, improvements in virtual slide scanning protocols would directly improve the quality of PMA-pretreated specimens subjected to CAIA

Objective: The aim of this study was to evaluate the effect of field of view (FOV) and beam energy on the scatter-to-primary ratio (SPR) in dental cone-beam CT (CBCT). Methods: An anthropomorphic phantom representing an adult male (ATOM Max 711-HN, Norfolk, VA, USA) was scanned using the 3D Accuitomo 170 CBCT (J. Morita, Kyoto, Japan) using 11 FOVs. During each scan, half of the X-ray beam was blocked. Each scan was performed at three exposure settings with varying beam energy and equal radiation dose: 90 kV 5 mA, 77 kV 7.5 mA and 69 kV 10 mA. The SPR was estimated by measuring the grey values in the blocked and non-blocked regions of the RAW data. The effect of FOV on SPR was evaluated using Dunn’s multiple comparison test, and the effect of the exposure settings was compared using a Wilcoxon signed rank test. Results: Larger FOVs showed increased scatter. FOVs with a shorter isocenter-detector distance showed a particularly high SPR. Most intercomparisons between FOVs were statistically significant. The largest difference was found between 17 × 12 cm and 6 × 6 cm (lower jaw), with the former showing a 4.9-fold higher SPR. The effect of beam energy was relatively small and varied between FOV sizes and positions. Conclusion: While the choice of FOV size and position is determined by the diagnostic region of interest, the image quality deterioration for large FOVs due to scatter provides another incentive to limit the FOV size as much as possible.

Background: Patients with Parkinson's disease (PD) and progressive supranuclear palsy Richardson's syndrome (PSP-RS) often show overlapping clinical features, leading to misdiagnoses. The objective of this study was to investigate the feasibility and utility of using multi-modal MRI datasets for an automatic differentiation of PD patients, PSP-RS patients, and healthy control (HC) subjects.Material and Methods: T1-weighted, T2-weighted, and diffusion-tensor (DTI) MRI datasets from 45 PD patients, 20 PSP-RS patients, and 38 HC subjects were available for this study. Using an atlas-based approach, regional values of brain morphology (T1-weighted), brain iron metabolism (T2-weighted), and microstructural integrity (DTI) were measured and employed for feature selection and subsequent classification using combinations of various established machine learning methods.Results: The optimal machine learning model using regional morphology features only achieved a classification accuracy of 65% (67/103 correct classifications) differentiating PD patients, PSP-RS patients, and HC subjects. The optimal machine learning model using only quantitative T2 values performed slightly better and achieved an accuracy of 75.7% (78/103). The optimal classifier using DTI features alone performed considerably better with 95.1% accuracy (98/103). The optimal multi-modal classifier using all features also achieved an accuracy of 95.1% but required more features and achieved a slightly lower F1-score compared to the optimal model using DTI features alone.Conclusion: Machine learning models using multi-modal MRI perform significantly better than uni-modal machine learning models using morphological parameters based on T1-weighted MRI datasets alone or brain iron metabolism markers based on T2-weighted MRI datasets alone. However, machine learnig models using regional brain microstructural integrity metrics computed from DTI datasets perform similar to the optimal multi-modal machine learning model. Thus, given the results from this study cohort, it appears that morphology and brain iron metabolism markers may not provide additional value for classification compared to using DTI metrics alone.

Objectives: The purpose of this study was to evaluate the effect of section thickness on volume estimations of bone defects scanned using cone beam computed tomography (CBCT). Methods: 25 bone defects were prepared on sheep mandibles and scanned using a KaVo 3D eXam (KaVo Dental, Biberach, Germany) CBCT device. Section thickness of images were reconstructed at 0.25, 0.5, and 0.75 mm to estimate the volume of these defects using the semiautomatic segmentation method. The volume averages obtained using microcomputed tomography and Archimedes’ method served as reference values. The estimated volumes at each section thickness were compared with the actual volumes using the Friedman test. The accuracy of volume estimation was determined by the percentage error with respect to the reference values, and the mean absolute error (MAE) was calculated. Results: Volumetric values of bone defects obtained with CBCT at section thicknesses up to 0.5 mm were compatible with the actual volumes (p > 0.05). The percentage errors at section thicknesses of 0.25, 0.5, and 0.75 mm were −5.4%, −7.3%, and −13.1%, respectively. The mean absolute errors were 13.6 mm3, 15.7 mm3, and 18.2 mm3, respectively. Conclusions: The section thickness values of CBCT images can be increased to a reasonable level to obtain accurate volume estimation results and save time. The semiautomatic segmentation method can be used reliably for volume estimations of bone defects.

The purpose of this study was to compare the accuracy of the planimetric methods on volume estimations by using cone beam computed tomography (CBCT). Thirty-one prepared intraosseous bone defects from thirteen bovine femur condyles were scanned with CBCT. The defect volumes were estimated by point counting (PC), manual segmentation (MS) and semiautomatic segmentation (SAS) methods at 0.3-mm section thickness without any intersection gap. The estimated volumes were compared with the results of the Archimedes’ method. The planimetric methods were analyzed using a Friedman’s two-way analysis of variance test. The estimated volumes of MS and SAS methods were compatible with the volumes of Archimedes’ method (p = 0.768, p = 0.140, respectively), but the volumes from the PC method were not compatible with Archimedes’ method (p

AIM To evaluate reproducibility of pulmonary embolism (PE) clot volume quantification using computed tomography pulmonary angiogram (CTPA) in a multicenter setting. METHODS This study was performed using anonymized data in conformance with HIPAA and IRB Regulations (March 2015-November 2016). Anonymized CTPA data was acquired from 23 scanners from 18 imaging centers using each site’s standard PE protocol. Two independent analysts measured PE volumes using a semi-automated region-growing algorithm on an FDA-approved image analysis platform. Total thrombus volume (TTV) was calculated per patient as the primary endpoint. Secondary endpoints were individual thrombus volume (ITV), Qanadli score and modified Qanadli score per patient. Inter- and intra-observer reproducibility were assessed using intra-class correlation coefficient (ICC) and Bland-Altman analysis. RESULTS Analyst 1 found 72 emboli in the 23 patients with a mean number of emboli of 3.13 per patient with a range of 0-11 emboli per patient. The clot volumes ranged from 0.0041 - 47.34 cm3 (mean +/- SD, 5.93 +/- 10.15cm3). On the second read, analyst 1 found the same number and distribution of emboli with a range of volumes for read 2 from 0.0041 – 45.52 cm3 (mean +/- SD, 5.42 +/- 9.53cm3). Analyst 2 found 73 emboli in the 23 patients with a mean number of emboli of 3.17 per patient with a range of 0-11 emboli per patient. The clot volumes ranged from 0.00459-46.29 cm3 (mean +/- SD, 5.91 +/- 10.06 cm3). Inter- and intra-observer variability measurements indicated excellent reproducibility of the semi-automated method for quantifying PE volume burden. ICC for all endpoints was greater than 0.95 for inter- and intra-observer analysis. Bland-Altman analysis indicated no significant biases. CONCLUSION Semi-automated region growing algorithm for quantifying PE is reproducible using data from multiple scanners and is a suitable method for image analysis in multicenter clinical trials.

Levels of essential metals in human breast milk (HBM) have been determined by different analytical techniques, but there is few woks about human whey milk fractions. However, the current trend lies in metalloproteomic and identification of different metalloproteins. In this sense, native separative techniques (N-PAGE and SEC) coupled to ICP-MS provide us with valuable information. Besides it is necessary the development of new methodologies in order to determine with accuracy and precision the profile of such metals and metalloproteins in the different whey protein fractions of HBM. Thus, the aim of this work was to develop a new method for metals and metalloproteins determination by SEC-ICP-MS in whey protein fractions of HBM. Human whey fractions were obtained of HBM samples by ultracentrifugation. Then, protein fractions of whey milk were separated by SEC coupled to ICP-MS for metalloproteins and Mn, Co, Cu and Se quantification. Besides, protein profile of whey milk was determined by N-PAGE and computer assisted image analysis. SEC-ICP-MS results indicated that first and second protein fractions showed detectable levels of the Mn, Co, Cu, and Se. Protein profile determined by N-PAGE and image analysis showed that molecular weight of protein fractions ranged between 68,878–1,228.277 Da. In this work, metalloproteins were analyzed by SEC coupled to ICP-MS, with adequate sensitivity and accuracy. Our study has shown the presence of Mn, Co, Cu and Se bound to two protein fractions in whey milk of HBM. Metals levels analyzed were within the ranges reported in the literature. Fil: Acosta, Mariano. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - San Luis. Instituto de Química de San Luis. Universidad Nacional de San Luis. Facultad de Química, Bioquímica y Farmacia. Instituto de Química de San Luis; Argentina. Universidad Nacional de San Luis. Facultad de Química, Bioquímica y Farmacia. Departamento de Química. Área de Química Analítica; Argentina Fil: Torres, Alberto Sabier. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - San Luis. Instituto de Química de San Luis. Universidad Nacional de San Luis. Facultad de Química, Bioquímica y Farmacia. Instituto de Química de San Luis; Argentina. Universidad Nacional de San Luis. Facultad de Química, Bioquímica y Farmacia. Departamento de Química. Área de Química Analítica; Argentina Fil: Marino Repizo, Leonardo. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - San Luis. Instituto de Química de San Luis. Universidad Nacional de San Luis. Facultad de Química, Bioquímica y Farmacia. Instituto de Química de San Luis; Argentina. Universidad Nacional de San Luis. Facultad de Química, Bioquímica y Farmacia. Departamento de Química. Área de Química Analítica; Argentina Fil: Martinez, Luis Dante. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - San Luis. Instituto de Química de San Luis. Universidad Nacional de San Luis. Facultad de Química, Bioquímica y Farmacia. Instituto de Química de San Luis; Argentina. Universidad Nacional de San Luis. Facultad de Química, Bioquímica y Farmacia. Departamento de Química. Área de Química Analítica; Argentina Fil: Gil, Raul Andres. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - San Luis. Instituto de Química de San Luis. Universidad Nacional de San Luis. Facultad de Química, Bioquímica y Farmacia. Instituto de Química de San Luis; Argentina. Universidad Nacional de San Luis. Facultad de Química, Bioquímica y Farmacia. Departamento de Química. Área de Química Analítica; Argentina

Background: The histological grading/staging system for Non-alcoholic steatohepatitis (NASH) encompasses the following criteria: extent of steatosis, lobular-inflammation, ballooning hepatocytes and extent of fibrosis. The existing scoring system has a drawback of high subjectivity. This study was designed with aims of studying (i) Inter & intra-observer variation; (ii) Correlation of the visual grading/staging with computer assisted image-analysis technique (IA); ultrasonography grade of fatty liver and fibroscan for extent of fibrosis. Methods:48 known cases of NASH were studied. The Sirius-red stained slides were assessed for % area of steatosis, fibrosis and parenchyma using Image Pro Plus® 6.1 image-analysis software. Results: The Kappa coefficient value for inter-observer variation was 0.338 (SE: 0.111-fair agreement); intra-observer variation was -0.083 (SE: 0.143-poor agreement). There was no agreement between the steatosis scoring between visual grading and IA score (K= -0.16). IA score unlike visual score correlated well with the USG grading system (K= 0.65) and biochemical variables [P values TG (0.04), BMI (0.039)]. For the first time in literature image analysis fibrosis score has been assessed in cases of NASH. The correlation between fibroscan score and IA fibrosis showed good correlation (r=0.766). The paper also lays down the cut off values of %fibrosis score on image analysis. A linear progression curve has been plotted to estimate extent of actual fibrosis in liver as against the existing fibroscan test. Conclusion: Visual assessment over-estimates the area of steatosis, resulting in false higher NASH histological grade, unlike the image-analysis system. The image-analysis adds objectivity to score steatosis and fibrosis.

AIM To examine effects of computed tomography (CT) image acquisition/reconstruction parameters on clot volume quantification in vitro for research method validation purposes. METHODS This study was performed in conformance with HIPAA and IRB Regulations (March 2015-November 2016). A ten blood clot phantom was designed and scanned on a dual-energy CT scanner (SOMATOM Force, Siemens Healthcare GmBH, Erlangen, Germany) with varying pitch, iterative reconstruction, energy level and slice thickness. A range of clot and tube sizes were used in an attempt to replicate in vivo emboli found within central and segmental branches of the pulmonary arteries in patients with pulmonary emboli. Clot volume was the measured parameter and was analyzed by a single image analyst using a semi-automated region growing algorithm implemented in the FDA-approved Siemens syngo.via image analysis platform. Mixed model analysis was performed on the data. RESULTS On the acquisition side, the continuous factor of energy showed no statistically significant effect on absolute clot volume quantification (P = 0.9898). On the other hand, when considering the fixed factor of pitch, there were statistically significant differences in clot volume quantification (P < 0.0001). On the reconstruction side, with the continuous factor of reconstruction slice thickness no statistically significant effect on absolute clot volume quantification was demonstrated (P = 0.4500). Also on the reconstruction side, with the fixed factor of using iterative reconstructions there was also no statistically significant effect on absolute clot volume quantification (P = 0.3011). In addition, there was excellent R2 correlation between the scale-measured mass of the clots both with respect to the CT measured volumes and with respect to volumes measure by the water displacement method. CONCLUSION Aside from varying pitch, changing CT acquisition parameters and using iterative reconstructions had no significant impact on clot volume quantification with a semi-automated region growing algorithm.

Background The increase in therapeutic choices in inflammatory bowel diseases (IBD) imposed the identification of personalized therapeutic strategy. Confocal laser endomicroscopy (CLE) is a new endoscopic tool developed to obtain virtual in vivo histology. This study aimed to identify CLE in vivo and ex vivo features predictive of response for patients starting biologics. Methods We performed a prospective observational study at a single tertiary referral centre. 29 IBD patients (14 ulcerative colitis-UC and 15 Crohn’s Disease-CD) underwent colonoscopy with CLE before and 12-14 weeks after starting anti -TNF or 16 weeks after anti α4β7 integrin. CLE parameters analyzed were: crypt distribution along the mucosal surface, crypt area (CA), eccentricity, diameter, inter-cryptic distance (ICD), vessel tortuosity (VT), fluorescein leakage through the colonic mucosa (FLCM). Ex-vivo binding activity of fluorescein labelled biologics on corresponding biopsies was also assessed. Mosaicism reconstruction of CLE images were analyzed using a dedicated software algorithm (CellvizioViewer, Mauna Kea Technologies, ParisFrance). A dedicated Graphical User Interface was designed to enable a semiautomated analysis. Results After treatment, VT resulted significantly changed in overall IBD population (p Conclusion FLCM and ICD were the best discriminants of response in UC, while VT, CA and ICD in CD. A higher mucosal binding to a biological agent before treatment was observed in responders UC patients but not in CD patients.

Background Biologics are being used increasingly in the treatment of Inflammatory Bowel Disease. However, up to 40% of patients do not respond to biologics. Therefore, methods to predict response are imperative. We aimed to identify novel genes and pathways predictive of anti-TNF response in patients with Ulcerative Colitis (UC) undergoing electronic chromoendoscopy and probe confocal laser endomicroscopy (pCLE). We further evaluated the ex-vivo binding of fluorescent labelled biologics as markers of response Methods 26 UC patients starting anti-TNF therapy as standard of care were recruited. Pre-treatment colonoscopy, with electronic chromoendoscopy and pCLE (Cellvizio, Mauna Kea) by injecting intravenous fluorescein (2.5-5mls), was performed to assess disease activity. Targeted biopsies were taken for fluorescein isothiocyanate (FITC)-labelled infliximab staining and RNA extraction and gene expression analysis. Ex vivo labelling was evaluated by an automated analysis: after a first pre-processing step to remove biases, the labelled regions were identified using statistical multi-level thresholding, and evaluated as area and intensity. To assess response, the same endoscopic procedure was repeated at week 12–14 after anti-TNF. cDNA libraries were prepared using QIAseq UPX 3’Transcriptome reagents and sequenced. Normalised gene expressions were obtained through the CLC Genomics Workbench. Differentially expressed genes (DEGs) (FDR-corrected P-value Results At baseline increased binding of the labelled biologic was associated with a higher likelihood of response to treatment (AUROC81%, accuracy77%, PPV100%, NPV63%). 342 DEGs (75 up-regulated, 267 down-regulated) distinguished responders from non-responders, 76 fell within enriched pathways. Pathways related to inflammation, chemotaxis, TGF-beta signalling, extracellular matrix and carbohydrate metabolism were reduced and cell-cell adhesion increased in responders pre-treatment. Among the 37 genes with VIP>1, CRIP2, CXCL6,EMILIN1,GADD45B, LAMA4 and MAPKAPK2 were upregulated in non-responders pre-treatment and were good predictors of response (AUROC>0.7) in this cohort and validation cohort Conclusion A higher mucosal binding of the biologics before treatment was observed in anti-TNF responders. Responsive UC patients have a less inflamed and fibrotic state pre-treatment. Chemotactic pathways, involving CXCL6 may be novel targets to treat non-responders

// Timothy Kwang Yong Tay 1 , Aye Aye Thike 1 , Nirmala Pathmanathan 1, 2 , Ana Richelia Jara-Lazaro 1, 3 , Jabed Iqbal 1 , Adeline Shi Hui Sng 1 , Heng Seow Ye 1 , Jeffrey Chun Tatt Lim 1 , Valerie Cui Yun Koh 1 , Jane Sie Yong Tan 1 , Joe Poh Sheng Yeong 1 , Zi Long Chow 1 , Hui Hua Li 4 , Chee Leong Cheng 1 and Puay Hoon Tan 5 1 Department of Anatomical Pathology, Singapore General Hospital, Singapore 2 Current affiliation: Westmead Breast Cancer Institute, Westmead Hospital, Westmead, NSW, Australia 3 Current affiliation: Q 2 Solutions – Central Laboratories, Singapore Science Park One, Singapore 4 Division of Medicine, Singapore General Hospital, Singapore 5 Division of Pathology, Singapore General Hospital, Singapore Correspondence to: Puay Hoon Tan, email: tan.puay.hoon@singhealth.com.sg Keywords: Ki67; breast cancer; computer assisted image analysis; prognosis Received: October 10, 2017 Accepted: January 25, 2018 Published: February 05, 2018 ABSTRACT Background: Ki67 positivity in invasive breast cancers has an inverse correlation with survival outcomes and serves as an immunohistochemical surrogate for molecular subtyping of breast cancer, particularly ER positive breast cancer. The optimal threshold of Ki67 in both settings, however, remains elusive. We use computer assisted image analysis (CAIA) to determine the optimal threshold for Ki67 in predicting survival outcomes and differentiating luminal B from luminal A breast cancers. Methods: Quantitative scoring of Ki67 on tissue microarray (TMA) sections of 440 invasive breast cancers was performed using Aperio ePathology ImmunoHistochemistry Nuclear Image Analysis algorithm, with TMA slides digitally scanned via Aperio ScanScope XT System. Results: On multivariate analysis, tumours with Ki67 ≥14% had an increased likelihood of recurrence (HR 1.941, p=0.021) and shorter overall survival (HR 2.201, p=0.016). Similar findings were observed in the subset of 343 ER positive breast cancers (HR 2.409, p=0.012 and HR 2.787, p=0.012 respectively). The value of Ki67 associated with ER+HER2-PR

Immunofluorescence (IF) and in situ proximity ligation assay (isPLA) are techniques that are used for in situ protein expression and colocalisation analysis, respectively. However, an efficient quantitative method to analyse both IF and isPLA staining on skin sections is lacking. Therefore, we developed a new method for semi-automatic quantitative layer-by-layer measurement of protein expression and colocalisation in skin sections using the free open-source software CellProfiler. As a proof of principle, IF and isPLA of ichthyosis-related proteins TGm-1 and SDR9C7 were examined. The results indicate that this new method can be used for protein expression and colocalisation analysis in skin sections.

Objectives: This study aimed to search for scientific evidence concerning the accuracy of computer-assisted analysis for diagnosing maxillofacial radiolucent lesions. Methods: A systematic review was conducted according to the statements of Preferred Reporting Items for Systematic Reviews and Meta-analyses Protocols and considering 10 databases, including the gray literature. Protocol was registered at the International Prospective Register of Systematic Reviews (CRD42018089945). The population, intervention, comparison and outcome strategy was used to define the eligibility criteria and only diagnostic test studies were included. Their risk of bias was assessed by the Joanna Briggs Institute Critical Appraisal tool. Random-effects model meta-analysis was performed and heterogeneity among the included studies was estimated using the I2 statistic. The grade of recommendation, assessment, development, and evaluation (GRADE) tool assessed the quality of evidence and strength of recommendation across included studies. Results: Out of 715 identified citations, four papers, published between 2009 and 2017, fulfilled the criteria and were included in this systematic review. A total of 191 lesions, classified as periapical granuloma and cyst, dentigerous cyst or keratocystic odontogenic tumor, were analyzed. All selected articles scored low risk of bias. The pooled accuracy estimation, regardless of the classification method used, was 88.75% (95% CI = 85.19-92.30). Heterogeneity test reached moderate values (I2 = 57.89%). According to the GRADE tool, the analyzed outcome was classified as having low level of certainty. Conclusions: The overall evaluation showed all studies presented high accuracy rates of computer-aided diagnosis systems in classifying radiolucent maxillofacial lesions compared to histopathological biopsy. However, due to the moderate heterogeneity found among the studies included in this meta-analysis, a pragmatic recommendation about the use of computer-assisted analysis is not possible.

Pulmonary hypertension (PH) is diagnosed invasively by right heart catheterization (RHC), which determines patient’s mean and systolic pulmonary artery pressure (mPAP, sPAP) and pulmonary vascular resistance (PVR). This study sought to identify non-invasive echocardiography parameters useful for screening PH. Patients ([Formula: see text]; 19 normotensive; 17 pre-capillary PH; 48 post-capillary PH) who had undergone transthoracic Doppler echocardiography and RHC within 60 days of each other were identified. Tricuspid regurgitant (TR) jet velocities, velocity spectral densities, average flow rates, and Fourier transforms (FFT) of velocity waveforms were calculated via an in-house MATLAB code. Correlations were found between the FFT magnitude at 0 Hz and sPAP and mPAP for normotensive patients; between the MATLAB-calculated TR jet and sPAP and PVR for all PH patients; and between the sum of FFT magnitudes [Formula: see text][Formula: see text]Hz and mPAP, sPAP, and PVR for post-capillary PH patients. Statistical difference was found between the FFT magnitudes at 2 Hz of pre- and post-capillary PH patients ([Formula: see text]). These results suggest non-invasive parameters with clinical utility for estimating RHC measurements and discriminating between PH types, offering a path forward for less invasive and more accessible PH screening protocol.

ResumenObjetivosEvaluar la adaptación de las mallas orbitarias preformadas industrialmente en nuestro entorno. Determinar la influencia de la planificación preoperatoria y la navegación en la precisión de la reconstrucción.Material y métodoSe estudió a pacientes con fracturas unilaterales de suelo orbitario reconstruidos con mallas de titanio preformadas entre 2009 y 2014. Las tomografías computarizadas (TC) se analizaron con iPlan 3.0 (BrainLab). Se importó la malla preformada en formato Standard Tesellation Language (STL) y se posicionó en la órbita sana reflejada en espejo sobre la fracturada. Se cuantificó la diferencia de volumen entre órbitas reconstruida y sana (DV), así como la adaptación del contorno de la malla a las superficies orbitarias. La navegación se hizo con el sistema Kolibrí (BrainLab).ResultadosSe incluyeron 17 reconstrucciones, 10 realizadas previa planificación y con navegación intraoperatoria. La DV fue significativamente menor en el grupo reconstruido con navegación (0,24±0,13 cc), p

The quality of 16 types of pork (PK) and poultry (PL) cold meats was evaluated by digital image analysis. Images were acquired in a flatbed scanner. The dry matter, protein, fat, ash and collagen content of the analyzed products was determined, and more than 2800 image texture variables from 12 color channels ( RGB , Lab* , XYZ , S , V , U ) were measured. The results were processed statistically by one-way ANOVA, correlation analysis, discriminant analysis and canonical analysis. Canonical analysis was performed to determine correlations between the chemical composition and image textures of cold meats. The developed statistical model discriminated cold meats with 89%–100% accuracy, subject to product type. The coefficients of correlation between chemical composition and image texture parameters were determined in the range of 0.70–0.92.

Morphometric analysis (measurements of size and shape) has become a significant research tool in phytolith studies. The International Phytolith Society (IPS) appointed the International Committee for Phytolith Morphometrics (ICPM) to establish methodological standards for the discipline. This paper presents current recommendations of the ICPM. It discusses the role of morphometric analysis in phytolith studies and recommends a paradigm for its application, criteria for data collection and publication, definitions for basic measurements and software for computer-assisted image analysis.

Summary Therapeutic decision-making in breast cancer depends on histopathologic biomarkers and is influenced by the Ki67 proliferation index. Computer-assisted image analysis (CAIA) promises to improve Ki67 quantification. Several commercial applications have been developed for semiautomated CAIA-based Ki67 quantification, many of which rely on measurements in user-defined regions of interest (ROIs). Because of intratumoral proliferative heterogeneity, definition of the ROI is an important step in the analytical procedure. This study explores the ROI size impacts on Ki67 quantification. Whole-slide sections of 100 breast cancers were immunostained with the anti–Ki67 antibody 30-9 and were analyzed on the iScan Coreo digital pathology platform using a Food and Drug Administration–cleared Ki67 quantification software version v5.3 (Virtuoso; Ventana, Tucson, TX). For each case, the Ki67 labeling index (LI) was determined in multiple ROIs of gradually increasing size centered around a high-proliferation area. The spatial Ki67 decline was modeled with nonlinear regression. Depending on the ROI size, the median Ki67 LI varied between 55% and 15%. The proportion of tumors classified as Ki67 low according to the St Gallen 2013/2015 cutoff increased from 2% to 56%, as the ROI size increased from 50 to 10 000 cells captured. The interrater reliability of conventional Ki67 assessment versus CAIA-based Ki67 quantification was also dependent on the ROI size and varied between slight and almost perfect agreement (Cohen κ = 0.06-0.85). In conclusion, the ROI size is a critically important parameter for semiautomated Ki67 quantification by CAIA. Ki67 LIs determined on platforms like iScan Coreo/Virtuoso require an ROI size adjustment, for which we offer a downloadable data transformation tool.

Mouse models of inflammatory bowel disease are critical for basic and translational research that is advancing the understanding and treatment of this disease. Assessment of these mouse models frequently relies on histologic endpoints. In recent years, whole slide imaging and digital pathology-based image analysis platforms have become increasingly available for implementation into the pathology workflow. These automated image analysis approaches allow for nonbiased quantitative assessment of histologic endpoints. In this study, the authors sought to develop an image analysis workflow using a commercially available image analysis platform that requires minimal training in image analysis or programming, and this workflow was used to score 2 mouse models of colitis that are primarily characterized by immune cell infiltrates in the lamina propria. Although the software was unable to accurately and consistently segment hematoxylin and eosin–stained sections, automated quantification of CD3 immunolabeling resulted in strong correlations with the pathologist’s score in all studies and allowed for the identification of 8 of the 9 differences among treatment groups that were identified by the pathologist. These results demonstrate not only the ability to incorporate solutions based on image analysis into the pathologist’s workflow but also the importance of immunohistochemical or histochemical surrogates for the incorporation of image analysis in histologic assessments.

Phenotypic traits are regularly used to diagnose the development of Fusarium head blight (FHB) in the field, whereas mycotoxin accumulation in wheat grains can only be accurately evaluated through costly methods, such as high–performance liquid chromatography (HPLC). The aim of this study was to determine whether: (i) the results provided by existing commercial decision support tools could be anticipated using phenotypic measurements, including a novel technique of computer–assisted image analysis of spikes; and (ii) these measurements could avoid using HPLC. We monitored the FHB development during two consecutive years in highly contaminated plots in the Burgundy region (France). Contamination by crop residues was simulated through a field inoculation with barley grains artificially colonized by Fusarium graminearum. The development of the disease on spikes and harvested grains was assessed on one tolerant and two susceptible wheat varieties. The accumulated amounts of mycotoxins were measured in harvested grains using HPLC. As expected, the measured traits revealed that the inoculum responsible for infection on spikes mainly came from residues left on the soil surface, and the susceptible varieties were more diseased than the tolerant variety. Weather conditions had a strong effect on disease development. The novel computer–assisted image analysis technique had a better prediction power of deoxynivalenol accumulation, was more objective and time–saving than classical visual symptom assessments. This assessment method could be suitable to supplement the use of existing prediction tools and might avoid systematic and costly mycotoxin measurements in likely infected plots.

Purpose To compare two measurement methods for body hair. Methods Calibration of computer assisted image analysis after manual processing (CAIAMP) showed variation

Delo prikazuje možnost uporabe analize slike za preučevanje mikroskopskih in makroskopskih objektov v gozdarski in lesarski stroki. Opiše strojno in programsko opremo za postavitev sistema za analizo slike in pove, kako dobimo sliko v računalnik. Ko je slika zajeta in shranjena na računalniški medij jo je potrebno pripraviti za nadaljnjo obdelavo. Najprej ji je potrebno izboljšati kakovost, nato izvesti klasifikacijo pikslov, umerjanje sistema in na koncu opraviti meritve. Meritve lahko opravljamo na klasificirani in neklasificirani sliki. Na koncu je podanih še nekaj teoretičnih in praktičnih primerov uporabe analize slike. Presented herewith are the possibilities of image analysis of micro- and macroobjects in wood technology and forestry. Described hardware and software needed to established an operational image analysis system. There is also described how to convert analogue pictures into a digital form and how to enhance image quality of the image. Image analysis is a five step process: (1)capturing the image, (2) enhancing the quality, (3) thresholding, (4) calibration of the system and (5) measuring. Measuring could be done on the original image for density studies or on thresholded images for different kinds of object analysis. Finally, some theoretical and practical examples arepresented. Key words

Although availability of automated platforms has proliferated, there is no standard practice for computer-assisted generation of scores for mRNA in situ hybridization (ISH) visualized by brightfield microscopic imaging on tissue sections. To address this systematically, an ISH for peptidylprolyl isomerase B (PPIB) (cyclophilin B) mRNA was optimized and applied to a tissue microarray of archival non-small cell lung carcinoma cases, and then automated image analysis for PPIB was refined across 4 commercially available software platforms. Operator experience and scoring results from ImageScope, HALO, CellMap, and Developer XD were systematically compared with each other and to manual pathologist scoring. Markup images were compared and contrasted for accuracy, the ability of the platform to identify cells, and the ease of visual assessment to determine appropriate interpretation. Comparing weighted scoring approaches using H-scores (Developer XD, ImageScope, and manual scoring) a correlation was observed (R value=0.7955), and association between the remaining 2 approaches (HALO and CellMap) was of similar value. ImageScope showed the highest R value in comparison with manual scoring (0.7377). Mean-difference plots showed that HALO produced the highest relative normalized values, suggesting higher relative sensitivity. ImageScope overestimated PPIB ISH signal at the high end of the range scores; however, this tendency was not observed in other platforms. HALO emerged with the highest number of favorable observations, no apparent systematic bias in score generation compared with the other methods, and potentially higher sensitivity to detect ISH. HALO may serve as a tool to empower teams of investigative pathology laboratory scientists to assist pathologists readily with quantitative scoring of ISH.

Optimization of treatment of the unilateral cleft lip nasal deformity (uCLND) is hampered by lack of objective means to assess initial severity and changes produced by treatment and growth. The purpose of this study was to develop automated 3D image analysis specific to the uCLND; assess the correlation of these measures to esthetic appraisal; measure changes that occur with treatment and differences amongst cleft types.Dorsum Deviation, Tip-Alar Volume Ratio, Alar-Cheek Definition, and Columellar Angle were assessed using computer-vision techniques. Subjects included infants before and after primary cleft lip repair (N = 50) and children aged 8-10 years with previous cleft lip (N = 50). Two expert surgeons ranked subjects according to esthetic nose appearance.Computer-based measurements strongly correlated with rankings of infants pre-repair (r = 0.8, 0.75, 0.41 and 0.54 for Dorsum Deviation, Tip-Alar Volume Ratio, Alar-Cheek Definition, and Columellar Angle, p 0.01) while all measurements except Alar-Cheek Definition correlated moderately with rankings of older children post-repair (r ∼ 0.35, p 0.01). Measurements were worse with greater severity of cleft type but improved following initial repair. Abnormal Dorsum Deviation and Columellar Angle persisted after surgery and were more severe with greater cleft type.Four fully-automated measures were developed that are clinically relevant, agree with expert evaluations and can be followed through initial surgery and in older children. Computer vision analysis techniques can quantify the nasal deformity at different stages, offering efficient and standardized tools for large studies and data-driven conclusions.

In the British Thoracic Society guidelines for incidental pulmonary nodules, volumetric analysis has become the recommended method for growth assessment in solid indeterminate pulmonary nodules (SIPNs) 3. In these guidelines, two different volume doubling time (VDT) cut-offs, 400 and 600 days, were proposed to differentiate benign from malignant nodules. The present study aims to evaluate the performance of these VDT cut-offs in a group of SIPNs 3 which were incidentally detected in a routine clinical setting. During a 7-year period, we retrospectively selected 60 patients with a single SIPN 3. For each SIPN, the volume and VDT were calculated using semiautomatic software throughout the follow-up period, and the performance of the 400- and 600-day VDT cut-offs was compared. In the selected sample, there were 38 benign and 22 malignant nodules. In this group of nodules, the sensitivity, negative predictive value and accuracy of the 600-day VDT cut-off were higher than those of the 400-day VDT cut-off. Therefore, in the management of SIPNs 3 which were incidentally detected in a clinical setting, the 600-day VDT cut-off was better at differentiating benign from malignant nodules than the 400-day VDT cut-off, by reducing the number of false negatives.

To develop a fully automated, accurate and robust segmentation technique for dental implants on cone-beam CT (CBCT) images.A head-size cylindrical polymethyl methacrylate phantom was used, containing titanium rods of 5.15 mm diameter. The phantom was scanned on 17 CBCT devices, using a total of 39 exposure protocols. Images were manually thresholded to verify the applicability of adaptive thresholding and to determine a minimum threshold value (Tmin). A three-step automatic segmentation technique was developed. Firstly, images were pre-thresholded using Tmin. Next, edge enhancement was performed by filtering the image with a Sobel operator. The filtered image was thresholded using an iteratively determined fixed threshold (Tedge) and converted to binary. Finally, a particle counting method was used to delineate the rods. The segmented area of the titanium rods was compared to the actual area, which was corrected for phantom tilting.Manual thresholding resulted in large variation in threshold values between CBCTs. After applying the edgeenhancing filter, a stable Tedge value of 7.5% was found. Particle counting successfully detected the rods for all but one device. Deviations between the segmented and real area ranged between -2.7 and +14.4mm(2) with an average absolute error of 2.8mm(2). Considering the diameter of the segmented area, submillimeter accuracy was seen for all but two data sets.A segmentation technique was defined which can be applied to CBCT data for an accurate and fully automatic delineation of titanium rods. The technique was validated in vitro and will be further tested and refined on patient data.

Resumen Objetivos: Evaluar la adaptación de las mallas orbitarias preformadas industrialmente en nuestro entorno. Determinar la influencia de la planificación preoperatoria y la navegación en la precisión de la reconstrucción. Material y método: Se estudió a pacientes con fracturas unilaterales de suelo orbitario reconstruidos con mallas de titanio preformadas entre 2009 y 2014. Las tomografías computarizadas (TC) se analizaron con iPlan 3.0 (BrainLab). Se importó la malla preformada en formato Standard Tesellation Language (STL) y se posicionó en la órbita sana reflejada en espejo sobre la fracturada. Se cuantificó la diferencia de volumen entre órbitas reconstruida y sana (DV), así como la adaptación del contorno de la malla a las superficies orbitarias. La navegación se hizo con el sistema Kolibrí (BrainLab). Resultados: Se incluyeron 17 reconstrucciones, 10 realizadas previa planificación y con navegación intraoperatoria. La DV fue significativamente menor en el grupo reconstruido con navegación (0,24 ± 0,13 cc), p < 0,01. En 9 pacientes la malla no se adaptaba adecuadamente en la TC preoperatoria, introduciéndose 1,88 ± 0,27 mm bajo el reborde infraorbitario y 3,23 ± 1,3 mm en la fosa nasal, con un ángulo medio de 13 ± 5,2°. La adaptación postoperatoria fue significativamente mejor en los casos en los que se había utilizado la navegación (p < 0,05). Conclusiones: Las mallas orbitarias preformadas necesitan ajustes en un 50% de nuestros casos. Planificación preoperatoria y navegación permiten identificar las órbitas en las que la adaptación no es buena y corregirla, y así mejorar la precisión de la reconstrucción. Abstract Objectives: To assess reconstruction with preformed orbital titanium meshes in our patients. To evaluate the influence of surgical planning and intraoperative navigation in orbital reconstruction accuracy. Material and method: Patients with unilateral orbital floor fractures reconstructed with preformed titanium meshes between 2009 and 2014 were included. Computed tomographies (CT) were analyzed with iPlan 3.0 (BrainLab). Orbital mesh was imported as a Standard Tesellation Language (STL) object and it was placed in the best position over the mirror uninjured orbit. Difference of volume between healthy and reconstructed orbits (VD) and variables to measure contour adaptation of the orbital mesh were evaluated. Intraoperative navigation was done with the BrainLab Kolibrí navigation system. Results: A total of 17 patients were reconstructed, 10 with preoperative planning and intraoperative navigation. VD was statistically lower in the group that was reconstructed using navigation (0.24 ± 0.13cc), P < .01. In 9 patients the position of the mesh was not adequate in the preoperative CT: the mesh protruded 1.88 ± 0.27 mm below the infraorbital rim, and 3.23 ± 1.3 mm in the nose with an angle of 13 ± 5.2°. Postoperative adaptation was statistically better in navigated patients (P < .05). Conclusions: Preformed orbital meshes needed adjustments in about 50% of our patients. Preoperative planning and surgical navigation help identifying them, increasing accuracy in their reconstructions.

Aim Navigation technology has given surgeons the capacity to know precisely where their instruments are during surgery, and simulation-guided navigation is a surgical method which is based on the use of navigation technology linked to 3D virtual surgery simulation. The aim of the present study was to evaluate the absolute accuracy of simulation-guided navigation as an aid to reproduce the planned position of the distraction device vector in pediatric mandibular distraction osteogenesis. Materials and methods We retrospectively evaluated seven patients affected by unilateral (3 pts.) or bilateral (4 pts.) mandibular hypoplasia and treated with this method between 2012 and 2014 at the Maxillofacial Surgery Unit of the University of Bologna, Italy. All patients were enrolled for mandibular distraction using internal unidirectional devices (11 distraction devices implanted overall). Patients were studied through a complete three-dimensional workflow, which led to obtaining a virtual reconstruction of the facial bones and the simulation of the distraction device positioning using a specific experimental software. The surgical planning was loaded on the navigation system and the distraction device was placed following the virtual plan, which is displayed on the navigation system as a guide and tracked with the navigation instruments. Results We analyzed the outcome comparing the recorded three-dimensional coordinates of the achieved distractor position and the three-dimensional coordinates of the planned distractor position. Among the eleven placed distractors, we found a mean angular error of 3.74° ± 3.30° on the axial axis (yaw) and of 6.27° ± 5.32° on the sagittal axis (pitch), while median angular errors are 3.72° on the axial axis (yaw) and 4.08° on the sagittal axis (pitch). Conclusion Our preliminary experience seems to support that simulation-guided navigation for vector control in mandibular distraction osteogenesis could be a useful procedure for reproducing the virtually planned outcome. Nevertheless, our results cannot be considered completely satisfactory. There is a lot of room to improve accuracy and errors seem to depend mainly on the soft tissues interference during distractor placement.

The aim of this study was to describe the temporal changes of various texture features extracted from pulmonary nodules on dynamic contrast-enhanced computed tomography (DCE-CT) and to compare the feature values among multiple scanning time points. We also aimed to analyze the variability of texture features across multiple scan delay times.This retrospective study was approved by the institutional review board of Seoul National University Hospital with waiver of patients' informed consent. Twenty patients (M:F, 6:14; mean age, 60.25 ± 11.97 years) with 20 lung nodules (mean size, 24.1 ± 12.3 mm) underwent DCE-CT with multiple scan delays (30, 60, 90, 120, 150, 180, 210, 240, 300, and 480 seconds) after precontrast scans. Lung nodule segmentation and texture feature extraction were performed at each time point using in-house software. Texture feature values were compared among the multiple time points using the Friedman test with post hoc pairwise Wilcoxon signed rank test. In addition, the dynamic range (DR) reflecting the variability between 2 time points to the interpatient range was calculated. Thereafter, we determined the stable time range that met both "DR greater than 0.90" and "no statistically significant difference" between all time point pairs for each feature. The degree of variability across all scan delay times was obtained using coefficients of variation.Standard deviation, variance, entropy, sphericity, discrete compactness, gray-level cooccurrence matrix (GLCM) inverse difference moment (IDM), GLCM contrast, and GLCM entropy did not show significant differences between scan delays of 30 and 180 seconds with DR greater than 0.90 between all time point pairs. When the range was narrowed down to 60 to 150 seconds, an additional 2 values (mean and homogeneity) showed stability. Among the 13 texture features, entropy, sphericity, discrete compactness, and GLCM entropy exhibited the lowest variability (coefficient of variation ≤5%).Most texture features exhibited stability with low variation between 60 and 150 seconds on DCE-CT. Thus, texture features extracted from contrast-enhanced CT with a scan delay range of 60 to 150 seconds can be used for tumor characterization despite the heterogeneity in delay time.