Automatic resin duct detection and measurement from wood core images using convolutional neural networks.

The structure and features of resin ducts provide valuable information about environmental conditions accompanying the growth of trees in the genus Pinus. Therefore analysis of resin duct characteristics has been an increasingly common measurement in dendrochronology. However, the measurement is tedious and time-consuming since it requires thousands of ducts to be manually marked in an image of an enlarged wood surface. Although tools exist to automate some stages of this process, no tool exists to automatically recognize and analyze the resin ducts and standardize them with the tree rings they belong to. This study proposes a new fully automatic pipeline that quantifies the properties of resin ducts in terms of the tree ring area to which they belong. A convolutional neural network underlays the pipeline to detect resin ducts and tree-ring boundaries. Also, a region merging procedure is used to identify connected components corresponding to successive rings. Corresponding ducts and rings are next related to each other. The pipeline was tested on 74 wood images representing five Pinus species. Over 8000 tree-ring boundaries and almost 25,000 resin ducts were analyzed. The proposed method detects resin ducts with a sensitivity of 0.85 and precision of 0.76. The corresponding scores for tree-ring boundary detection are 0.92 and 0.99, respectively.

Image processing pipeline for the detection of blood flow through retinal vessels with subpixel accuracy in fundus images.

BACKGROUND AND OBJECTIVE: Blood flow detection through the retinal vessels is an essential step in diagnosing several eye diseases. It manifests itself as changes in vessel diameter in consecutive phases of blood flow. Previous studies focused mainly on determining retinal vessel diameter by the manual demarcation of vessel edges, which is time-consuming. As a result, only a few selected vessels were considered, which is not reliable. Such measurements are also prone to human errors and operator subjectivity, which additionally decreases their reliability. For these reasons, this paper proposes an automated pipeline to analyze the blood flow through retinal vessels. METHODS: Convolutional neural networks were used for optic disc and vessel detection and full width at half maximum analysis used for vessel width assessment at the subpixel level. Measurements of the diameter were performed for five phases of the blood flow to all vessels at a particular distance derived from the optic disc size. We tested the approach on fundus images of five patients, with both eyes examined in each participant. The threshold for the detections of blood flow was when the retinal diameter vessel measurements were above the camera's resolution as compared among all 5 phases of blood flow. RESULTS: A total of 205 large caliber blood vessels were analyzed with blood flow detected in 18 retinal blood vessels. Conclusions Average vessel diameters derived from manual and automatic measurements differed on average by 4.96%. Average relative errors for single vessel measurements along the vessels range from 4.21 to 11.85%, with a global average at the level of 8%. Therefore, the measurements can be considered as accurate and in a high agreement between the expert and algorithm.

Segmentation of corneal endothelium images using a U-Net-based convolutional neural network.

Diagnostic information regarding the health status of the corneal endothelium may be obtained by analyzing the size and the shape of the endothelial cells in specular microscopy images. Prior to the analysis, the endothelial cells need to be extracted from the image. Up to today, this has been performed manually or semi-automatically. Several approaches to automatic segmentation of endothelial cells exist; however, none of them is perfect. Therefore this paper proposes to perform cell segmentation using a U-Net-based convolutional neural network. Particularly, the network is trained to discriminate pixels located at the borders between cells. The edge probability map outputted by the network is next binarized and skeletonized in order to obtain one-pixel wide edges. The proposed solution was tested on a dataset consisting of 30 corneal endothelial images presenting cells of different sizes, achieving an AUROC level of 0.92. The resulting DICE is on average equal to 0.86, which is a good result, regarding the thickness of the compared edges. The corresponding mean absolute percentage error of cell number is at the level of 4.5% which confirms the high accuracy of the proposed approach. The resulting cell edges are well aligned to the ground truths and require a limited number of manual corrections. This also results in accurate values of the cell morphometric parameters. The corresponding errors range from 5.2% for endothelial cell density, through 6.2% for cell hexagonality to 11.93% for the coefficient of variation of the cell size.

A novel approach for quantification of time-intensity curves in a DCE-MRI image series with an application to prostate cancer.

This paper considers the problem of an automatic quantification of DCE-MRI curve shape patterns. In particular, the semi-quantitative approach which classifies DCE time-intensity curves into clusters representing the tree main shape patterns is proposed. The approach combines heuristic rules with the naive Bayes classifier. In particular, the descriptive parameters are firstly derived from pixel-by-pixel analysis of the DCE time intensity curves and then used to recognise the curves which without a doubt represent the three main shape patterns. These curves are next used to train the naive Bayes classifier intended to classify the remaining curves within the dataset. Results of applying the proposed approach to the DCE-MRI scans of patients with prostate cancer are presented and discussed. Additionally, the overall performance of the approach is estimated through the comparison with the ground truth results provided by the expert.

The quantitative assessment of the pre- and postoperative craniosynostosis using the methods of image analysis.

This paper considers the problem of the CT based quantitative assessment of the craniosynostosis before and after the surgery. First, fast and efficient brain segmentation approach is proposed. The algorithm is robust to discontinuity of skull. As a result it can be applied both in pre- and post-operative cases. Additionally, image processing and analysis algorithms are proposed for describing the disease based on CT scans. The proposed algorithms automate determination of the standard linear indices used for assessment of the craniosynostosis (i.e. cephalic index CI and head circumference HC) and allow for planar and volumetric analysis which so far have not been reported. Results of applying the introduced methods to sample craniosynostotic cases before and after the surgery are presented and discussed. The results show that the proposed brain segmentation algorithm is characterized by high accuracy when applied both in the pre- and postoperative craniosynostosis, while the introduced planar and volumetric indices for the disease description may be helpful to distinguish between the types of the disease.

The segmentation of 3D images using the random walking technique on a randomly created image adjacency graph.

This paper considers the problem of image segmentation using the random walker algorithm. In the case of 3D images, the method uses an extreme amount of memory and time resources. These are required in order to represent the corresponding enormous image graph and to solve the resulting sparse linear system. Having in mind these limitations, this paper proposes techniques for the optimization of the random walker approach. The optimization is obtained by processing supervoxels representing homogeneous image regions rather than single voxels. A fast and efficient method for supervoxel determination is introduced. A method for the creation of an image adjacency graph from an irregular grid of supervoxels is also proposed. The results of applying the introduced approach to segmentation of 3D CT data sets are presented and compared with the results of the original random walker approach and other state-of-the-art methods. The accuracy and the computational overhead is regarded in the comparison. The analysis of results shows that the modified method can be successfully applied for the segmentation of volumetric images and provides results in a reasonable time without a significant loss in the image segmentation accuracy. It also outperforms the state-of-the-art methods considered in the comparison.

Comparing algorithms for automated vessel segmentation in computed tomography scans of the lung: the VESSEL12 study.

The VESSEL12 (VESsel SEgmentation in the Lung) challenge objectively compares the performance of different algorithms to identify vessels in thoracic computed tomography (CT) scans. Vessel segmentation is fundamental in computer aided processing of data generated by 3D imaging modalities. As manual vessel segmentation is prohibitively time consuming, any real world application requires some form of automation. Several approaches exist for automated vessel segmentation, but judging their relative merits is difficult due to a lack of standardized evaluation. We present an annotated reference dataset containing 20 CT scans and propose nine categories to perform a comprehensive evaluation of vessel segmentation algorithms from both academia and industry. Twenty algorithms participated in the VESSEL12 challenge, held at International Symposium on Biomedical Imaging (ISBI) 2012. All results have been published at the VESSEL12 website http://vessel12.grand-challenge.org. The challenge remains ongoing and open to new participants. Our three contributions are: (1) an annotated reference dataset available online for evaluation of new algorithms; (2) a quantitative scoring system for objective comparison of algorithms; and (3) performance analysis of the strengths and weaknesses of the various vessel segmentation methods in the presence of various lung diseases.

[The prevalence of tinnitus in children in Poland]. / Ocena czestosci wystepowania szumów usznych u dzieci w Polsce.

AIM: The objective of this study was to estimate the prevalence of the tinnitus in 7- and 12-year-old children in Poland. MATERIAL AND METHODS: In this study 118,005, 7-year-old children and 23,339, 12-year-old children and their parents were asked about the presence of tinnitus by answering a questionnaire. 7-year-old children were asked by tester if they can hear noise in their ears or head. RESULTS: The results showed that according to parents questionnaire answers tinnitus was present in group of 15,244 (12.9%) 7-year-old children and in group of 3886 (16.6%) 12-year-old children. Between 7-year-old children directly asked for tinnitus 29.3% (34,517) mentioned of having it. Group of 8060 (34.5%) 12-year-old children confirmed in questionnaire hearing tinnitus. Differences in the presence of tinnitus between 7- and 12-year-old children have proven to be statistically significant. CONCLUSIONS: Tinnitus is frequent complain among 7- and 12-year-old children. It is recommended to include to a questionnaire an inquiry about the presence of tinnitus during hearing screening tests performed recently more frequently at primary schools for early detection a diagnosis of tinnitus.

The relationship between distortion product otoacoustic emissions and extended high-frequency audiometry in tinnitus patients. Part 1: normally hearing patients with unilateral tinnitus.

BACKGROUND: The aim of this study was to evaluate distortion product otoacoustic emissions (DPOAEs) and extended high-frequency (EHF) thresholds in a control group and in patients with normal hearing sensitivity in the conventional frequency range and reporting unilateral tinnitus. MATERIAL/METHODS: Seventy patients were enrolled in the study: 47 patients with tinnitus in the left ear (Group 1) and 23 patients with tinnitus in the right ear (Group 2). The control group included 60 otologically normal subjects with no history of pathological tinnitus. Pure-tone thresholds were measured at all standard frequencies from 0.25 to 8 kHz, and at 10, 12.5, 14, and 16 kHz. The DPOAEs were measured in the frequency range from approximately 0.5 to 9 kHz using the primary tones presented at 65/55 dB SPL. RESULTS: The left ears of patients in Group 1 had higher median hearing thresholds than those in the control subjects at all 4 EHFs, and lower mean DPOAE levels than those in the controls for almost all primary frequencies, but significantly lower only in the 2-kHz region. Median hearing thresholds in the right ears of patients in Group 2 were higher than those in the right ears of the control subjects in the EHF range at 12.5, 14, and 16 kHz. The mean DPOAE levels in the right ears were lower in patients from Group 2 than those in the controls for the majority of primary frequencies, but only reached statistical significance in the 8-kHz region. CONCLUSIONS: Hearing thresholds in tinnitus ears with normal hearing sensitivity in the conventional range were higher in the EHF region than those in non-tinnitus control subjects, implying that cochlear damage in the basal region may result in the perception of tinnitus. In general, DPOAE levels in tinnitus ears were lower than those in ears of non-tinnitus subjects, suggesting that subclinical cochlear impairment in limited areas, which can be revealed by DPOAEs but not by conventional audiometry, may exist in tinnitus ears. For patients with tinnitus, DPOAE measures combined with behavioral EHF hearing thresholds may provide additional clinical information about the status of the peripheral hearing.

[The influence of high frequency hearing loss on the distortion product otoacoustic emissions in tinnitus subjects with normal hearing threshold (0,25-8 kHz)]. / Wplyw ubytków sluchu w zakresie wysokich czestotliwosci na otoemisje produktów znieksztalcen nieliniowych slimaka u osób z szumami usznymi i prawidlowym progiem slyszenia w zakresie 0,25-8kHz.

AIM OF THE STUDY: To evaluate the influence of high frequency hearing loss (> 8000 Hz) on distortion product otoacoustic emissions registered in the frequency range from 0,5 to 8 kHz. MATERIAL AND METHODS: 280 ears with tinnitus and normal hearing (0.25-8 kHz) divided into 3 groups depending on the degree of high frequency hearing loss: group A--hearing threshold up to 20 dB for 10, 12.5, 14 and 16 kHz (68 ears); group B--hearing threshold 25-40 dB HL for at least one of four EHfs (93 ears); group C--hearing threshold above 40 dB HL for at least one of four EHFs (119 ears). For each group mean audiogram and DP-gram were obtained and statistical analysis was used for comparison across these groups. RESULTS: Mean DPOAE values in group C were significantly lower in comparison with group A for the frequency range 2-8 kHz, and in comparison with group B were significantly lower for the frequency range 4-8 kHz. CONCLUSIONS: High frequency hearing loss (above 8 kHz) has a relevant influence on distortion product otoacoustic emissions registered at frequencies below 8 kHz in tinnitus patients. The greater is hearing loss above 8 kHz, the lower is DPOAE value below 8 kHz.

Extraction of airways from CT (EXACT'09).

This paper describes a framework for establishing a reference airway tree segmentation, which was used to quantitatively evaluate fifteen different airway tree extraction algorithms in a standardized manner. Because of the sheer difficulty involved in manually constructing a complete reference standard from scratch, we propose to construct the reference using results from all algorithms that are to be evaluated. We start by subdividing each segmented airway tree into its individual branch segments. Each branch segment is then visually scored by trained observers to determine whether or not it is a correctly segmented part of the airway tree. Finally, the reference airway trees are constructed by taking the union of all correctly extracted branch segments. Fifteen airway tree extraction algorithms from different research groups are evaluated on a diverse set of twenty chest computed tomography (CT) scans of subjects ranging from healthy volunteers to patients with severe pathologies, scanned at different sites, with different CT scanner brands, models, and scanning protocols. Three performance measures covering different aspects of segmentation quality were computed for all participating algorithms. Results from the evaluation showed that no single algorithm could extract more than an average of 74% of the total length of all branches in the reference standard, indicating substantial differences between the algorithms. A fusion scheme that obtained superior results is presented, demonstrating that there is complementary information provided by the different algorithms and there is still room for further improvements in airway segmentation algorithms.

[The prevalence of tinnitus in 7-year-old children in the eastern of Poland]. / Ocena czestosci wystepowania szumów usznych wsród 7-latków w Polsce Wschodniej.

INTRODUCTION: Tinnitus in children is still rarely mentioned in medical literature. The prevalence of pediatric tinnitus has been reported to range between 3-37% and higher in children with hearing loss. OBJECTIVES: The objective of this study was to estimate the prevalence of the tinnitus in 7 years old children and to conduct statistical analysis of the relation between tinnitus and hearing loss. MATERIAL AND METHODS: In this study 60 212 children and their parents were asked about the presence of tinnitus by answering a questionnaire at home. Additionally, children were asked if they can hear noise in their ears or head, before hand hearing screening test at school was conducted. The hearing screening test was performed using screening audiometer with air conduction. RESULTS: The results showed that tinnitus was present in group of 8164 children (13.6%). Between children directly asked for tinnitus, 33% mentioned of having it. Over 75% of children did not complained of tinnitus spontaneously to parents. Children with hearing loss reported tinnitus twice more comparing to children with normal hearing. CONCLUSIONS: Tinnitus is frequent complain among 7 years old children. Children with hearing loss reported tinnitus twice more comparing to children with normal hearing. It is recommended to include to a questionnaire an inquiry about the presence of tinnitus during hearing screening tests performed recently more frequently at primary schools for early detection and diagnosis of tinnitus.

[Distortion product otoacoustic emission levels and input/output-growth functions in normal-hearing individuals with tinnitus and/or hyperacusis]. / Otoemisja produktów znieksztalcen nieliniowych slimaka oraz funkcja wejscia/wyjscia u prawidlowo slyszacych pacjentów z szumem usznym i/lub nadwrazliwoscia sluchowa.

The purpose of this study was to determine whether distortion product otoacoustic emissions (DPOAEs) can be used to distinguish among four groups with audiometrically normal-hearing sensitivity: (1) control adults without tinnitus or hyperacusis, (2) patients with tinnitus alone, (3) patients with hyperacusis alone, and (4) patients with both tinnitus and hyperacusis. Two types of DPOAE measures were evaluated: (1) the distortion product (DP-)gram measured with fixed primary levels as a function of frequency, and (2) DPOAE input/output (I/O) functions for a range of primary levels between 45 and 70 dB sound pressure level (SPL) at 1000, 2000, and 4000 Hz. DP-grams did not clearly distinguish between the control and patient groups. There was, however, a consistent trend for the three patient groups to have decreased average DP levels at 4000 and 6000 Hz; this notch in the DP-gram was not observed in the response configuration for the control group. In the three patient groups, 51 to 74% of these individuals had DP levels that were outside of the 95% confidence range for the control group. The average slopes of the I/O growth functions for each of the patient groups were consistently steeper than those for the control group; however, the slope values were indistinguishable among the patient groups. About 60% of the patients' DPOAE responses (in each group) were categorized as abnormal based on their slope values. Thus, DPOAE measures can be used with at least partial success to distinguish controls from patients with tinnitus, hyperacusis, or both tinnitus and hyperacusis, but not to discriminate among the respective patient groups. These findings suggest that the pathology represented among the patient groups is consistent at the level of the cochlea; however, diagnostic tests targeted at higher centers of processing are needed if the individuals in these groups are to be distinguished among themselves. In all participants, DPOAEs with fixed primary levels as a function of frequency were measured. There was a consistent trend for the three patient groups to have decreased average DP-gram levels at 4000 and 6000 Hz; this notch in the DP-gram was not observed in the response configuration for the control group. In the three patient groups, 51 to 74% of these individuals had DP levels that were outside of the 95% confidence range for the control group. Thus, DPOAE measures can be used with at least partial success to distinguish controls from patients with tinnitus, hyperacusis, or both tinnitus and hyperacusis, but not to discriminate among the respective patient groups. These findings suggest that the pathology represented among the patient groups is consistent at the level of the cochlea; however, diagnostic tests targeted at higher centers of processing are needed if the individuals in these groups are to be distinguished among themselves.

Two-pass region growing algorithm for segmenting airway tree from MDCT chest scans.

The paper addresses problem of pulmonary airways investigation based on high-resolution multidetector computed tomography (MDCT) chest scans. Especially it presents new, fully automated algorithm for airway tree segmentation. The algorithm uses two passes of 3D seeded region growing. First pass is applied for obtaining the initial (rough) airway tree. The second pass aims at refining the tree based on the morphological gradient. Results of applying proposed algorithm to scans of several randomly selected patients are introduced and discussed. Moreover, comparison with results obtained by simple region growing with manual threshold selection is provided. Obtained results justify the method and prove that it detects up to 10 generations of bronchi and diminishes leakages into the lung parenchyma which are common when simple region growing is used for segmenting airway tree.

[Tinnitus in young patients up to 35-years old]. / Szumy uszne u osób mlodych do 35 roku zycia.

According to our observations tinnitus is becoming increasingly common amongst younger persons. About 25% of patients registered in our Tinnitus Clinic, are below the age of 35. Group of 235 tinnitus patients ranging in age from 18 to 35 years old were evaluated for this study. There were no other pre-selection criteria except the age. All of the patients answered the questions concerned with the cause of tinnitus. The results of our research indicate that infections (68.5%), noise exposure (27.7%) and stress (23.4%) are the most frequent cause of tinnitus among patients below 35 years.

[Analysis of the distortion product otoacoustic emission (DPOAE) and input/output function (I/O) in tinnitus patient with normal hearing]. / Analiza sygnalu i funkcji wzrostu otoemisji produktów znieksztalcen nieliniowych slimaka u pacjentów z szumem usznym i prawidlowa czuloscia sluchu.

Distortion Product Otoacoustic Emission (DPOAE) has been used as a tool to determine the presence of cochlear pathology. It is well known that the damage of Outer Hair Cells (OHCs) results in reduced auditory sensitivity and generation of tinnitus. At the same time tinnitus arises in normal hearing persons as well. The aim of this study was the assessment of the DPOEA amplitude and the behavior of DP I/O function in tinnitus patients with normal hearing threshold. The group of 48 tinnitus patients (92 ears) with normal hearing threshold and the control group of 30 normal hearing patients (40 ears) were investigated for this study. All subjects have undergone the measurement of DPOAE and DP I/O functions for frequencies: 1000, 2000 and 4000 Hz. In the tinnitus group there were observed the decrease in DPOAE amplitude for some frequencies and the steeper slope of I/O function. This means that the places with localized damage or dysfunction of OHCs on the basilar membrane can be responsible for the origin of tinnitus.