Breathing Phases Research Articles

Choosing a reference phase for a dynamic tumor tracking treatment: A new degree of freedom?

With the introduction of dynamic tumor tracking in radiotherapy, it is possible to irradiate moving targets with minimal safety margins. However, most dynamic tumor tracking techniques rely on changing the beam geometry by, for example, adapting the multileaf collimator (MLC) positions or rotating the LINAC head. These changes are relative to a reference position which is determined by a specific breathing phase. Since these changes in the beam path also influence the delivered dose, choosing a different reference position based on a different breathing phase impacts the applied dose to the patient. This work investigates the influence of choosing different reference breathing phases on the dose distribution. The Vero system tracks the moving target by performing a pan and tilt rotation of the LINAC head. For 13 patients, the target position was extracted from every phase of a four-dimensional computed tomography (4DCT) and the pan and tilt values were determined with respect to three different reference phases. These reference phases were inspiration, expiration, and the midventilation. For all reference phases, a 4D dose calculation was performed on the 4DCT regarding the respective pan and tilt values. Furthermore, the applied dose to the target and surrounding organs at risk was calculated. To accumulate the dose distribution, weights from the actual patient breathing motion were determined. The weights were calculated from the breathing motions from different days to investigate the impact of daily variations in the breathing motion onto the accumulated dose distribution. All obtained values were then compared to the static treatment plan. The mean and maximum doses applied to the target or surrounding organs at risk show no general behavior depending on the different reference phases. Nevertheless, for some patients, large differences (approx. 30%) in the applied dose to certain organs at risk could be observed, whereas the applied dose to the target shows no dependency on the different reference phases. However, the mean target dose is in all cases approx. 1.5% below the reference value from the static treatment plan. Although no general dependency of the applied dose on the selected reference phase could be found, the choice of the reference phase can have great impact on the organ at risk dose for some patients. Thus, the choice of the reference phase used for patient positioning should be considered during treatment planning since it can be seen as a new degree of freedom of a treatment based on tracking.

Breathing deformation model - application to multi-resolution abdominal MRI.

Dynamic MRI is a technique of acquiring a series of images continuously to follow the physiological changes over time. However, such fast imaging results in low resolution images. In this work, abdominal deformation model computed from dynamic low resolution images have been applied to high resolution image, acquired previously, to generate dynamic high resolution MRI. Dynamic low resolution images were simulated into different breathing phases (inhale and exhale). Then, the image registration between breathing time points was performed using the B-spline SyN deformable model and using cross-correlation as a similarity metric. The deformation model between different breathing phases were estimated from highly undersampled data. This deformation model was then applied to the high resolution images to obtain high resolution images of different breathing phases. The results indicated that the deformation model could be computed from relatively very low resolution images.

Distinct pathways to the parafacial respiratory group to trigger active expiration in adult rats.

Active expiration (AE) is part of the breathing phase; it is conditional and occurs when we increase our metabolic demand, such as during hypercapnia, hypoxia, or exercise. The parafacial respiratory group (pFRG) is involved in AE. Data from the literature suggest that excitatory and the absence of inhibitory inputs to the pFRG are necessary to determine AE. However, the source of the inputs to the pFRG that trigger AE remains unclear. We show in adult urethane-anesthetized Wistar rats that the pharmacological inhibition of the medial aspect of the nucleus of the solitary tract (mNTS) or the rostral aspect of the pedunculopontine tegmental nucleus (rPPTg) is able to generate AE. In addition, direct inhibitory projection from the mNTS or indirect cholinergic projection from the rPPTg is able to contact pFRG to trigger AE. The inhibition of the mNTS or the rPPTg under conditions of high metabolic demand, such as hypercapnia (9-10% CO2), did not affect the AE. The present results suggest for the first time that inhibitory sources from the mNTS and a cholinergic pathway from the rPPTg, involving M2/M4 muscarinic receptors, could be important sources to modulate and sustain AE.

Pengaruh slow deep breathing terhadap nyeri pada pasien post op apendisitis di rsud sleman

Slow Deep Breathing (SDB) is a breathing technique that can be used below 10 times per minute with a long phase of breathing. Objective: Effects of Slow Respiratory In the decrease of pain level in RSUD Sleman Method: Experimental design with pre-post test type without control group design. Samples of 30 respondents with Acidental Sampling Result: There is effect of Slow Deep Breathing on the pain decrease p 0,001@ 0,05 where P <0,05 can be concluded that there is significant difference. Conclusion: Slow Respiratory Exercise Slowly Reduces Pain Rate in Post-Appendicitis Patients at RSUD Sleman

Identification of asthma severity levels through wheeze sound characterization and classification using integrated power features

ObjectiveThis study aimed to investigate and classify wheeze sound characteristics according to asthma severity levels (mild, moderate and severe) using integrated power (IP) features. MethodValidated and segmented wheeze sounds were obtained from the lower lung base (LLB) and trachea recordings of 55 asthmatic patients with different severity levels during tidal breathing manoeuvres. From the segments, nine datasets were obtained based on the auscultation location, breath phases and their combination. In this study, IP features were extracted for assessing asthma severity. Subsequently, univariate and multivariate (MANOVA) statistical analyses were separately implemented to analyse behaviour of wheeze sounds according to severity levels. Furthermore, the ensemble (ENS), k-nearest- neighbour (KNN) and support vector machine (SVM) classifiers were applied to classify the asthma severity levels. Results and conclusionThe univariate results of this study indicated that the majority of features significantly discriminated (p < 0.05) the severity levels in all the datasets. The MANOVA results yielded significantly (p < 0.05) large effect size in all datasets (including LLB-related) and almost all post hoc results were significant (p < 0.05). A comparison of the performance of classifiers revealed that eight of the nine datasets showed improved performance with the ENS classifier. The Trachea inspiratory (T-Inspir) dataset produced the highest performance. The overall best positive predictive rate (PPR) for the mild, moderate and severe severity levels were 100% (KNN), 92% (SVM) and 94% (ENS) respectively. Analysis related to auscultation locations revealed that tracheal wheeze sounds are more specific and sensitive predictors of asthma severity. Additionally, phase related investigations indicated that expiratory and inspiratory wheeze sounds are equally informative for the classification of asthma severity.

Calculation of Inter- and Intra-Fraction Motion Errors at External Radiotherapy Using a Markerless Strategy Based on Image Registration Combined with Correlation Model

Introduction: A new method based on image registration technique and an intelligent correlation model to calculate. The present study aimed to propose inter- and intra-fraction motion errors in order to address the limitations of conventional Patient positioning methods. Material and Methods: The configuration of the markerless method was accomplished by using four-dimensional computed tomography (4DCT) datasets. Firstly, the MeVisLab software package was used to extract a three-dimensional (3D) surface model of the patient and determine the tumor location. Then, the patient-specific 3D surface model which also included the breathing phases was imported into the MATLAB software package in order to define several control points on the thorax region as virtual external markers. Finally, based on the correlation of breathing signals/patient position with breathing signals/tumor coordinate, an adaptive neuro fuzzy inference system was proposed to both verify and align the inter- and intra-fraction motion errors in radiotherapy, if needed. In order to validate the proposed method, the 4DCT data acquired from four real patients was considered. Results: Final results revealed that our hybrid configuration method was capable of aligning patient setup with lower uncertainties, compared to other available methods. In addition, the 3D root-mean-square error has been reduced from 5.26 to 1.5 mm for all patients. Conclusion: In this study, a markerless method based on the image registration technique in combination with a correlation model was proposed to address the limitations of the available methods, including dependence on operator’s attention, use of passive markers, and rigid-only constraint for patient setup.

Convolutional Neural Network for Breathing Phase Detection in Lung Sounds.

We applied deep learning to create an algorithm for breathing phase detection in lung sound recordings, and we compared the breathing phases detected by the algorithm and manually annotated by two experienced lung sound researchers. Our algorithm uses a convolutional neural network with spectrograms as the features, removing the need to specify features explicitly. We trained and evaluated the algorithm using three subsets that are larger than previously seen in the literature. We evaluated the performance of the method using two methods. First, discrete count of agreed breathing phases (using 50% overlap between a pair of boxes), shows a mean agreement with lung sound experts of 97% for inspiration and 87% for expiration. Second, the fraction of time of agreement (in seconds) gives higher pseudo-kappa values for inspiration (0.73–0.88) than expiration (0.63–0.84), showing an average sensitivity of 97% and an average specificity of 84%. With both evaluation methods, the agreement between the annotators and the algorithm shows human level performance for the algorithm. The developed algorithm is valid for detecting breathing phases in lung sound recordings.

Solvent‐Induced Control over Breathing Behavior in Flexible Metal–Organic Frameworks for Natural‐Gas Delivery

AbstractFinding appropriate stimuli for controlling the breathing behavior of flexible metal–organic frameworks (MOFs) is highly challenging. Herein, we report the solvent‐induced changes in the particle size and stability of different breathing phases of the MIL‐53 series, a group of flexible MOFs. A water/dimethylformamide (DMF) ratio is tuned to synthesize members of the MIL‐53 series which have different behaviors. The breathing is explored by high‐pressure methane sorption tests. Increasing DMF concentration decreases MOF particle size and increases the stability of the porous phases, boosting the 5.8–65 bar sorption difference of methane, which is required for natural‐gas delivery.

Solvent-Induced Control over Breathing Behavior in Flexible Metal-Organic Frameworks for Natural-Gas Delivery.

Finding appropriate stimuli for controlling the breathing behavior of flexible metal-organic frameworks (MOFs) is highly challenging. Herein, we report the solvent-induced changes in the particle size and stability of different breathing phases of the MIL-53 series, a group of flexible MOFs. A water/dimethylformamide (DMF) ratio is tuned to synthesize members of the MIL-53 series which have different behaviors. The breathing is explored by high-pressure methane sorption tests. Increasing DMF concentration decreases MOF particle size and increases the stability of the porous phases, boosting the 5.8-65 bar sorption difference of methane, which is required for natural-gas delivery.

Life and scientific creativity of professor A.E. Prozorov (to the 130 anniversary since birth)

The main milestones of life and scientific activity of professor A.E. Prozorov are provided. It is noted that Alexander Evgrafovich organized the first X-ray radiotherapeutic office at Gynecologic institute. From 1918 to 1924 served in the Red Army as the radiologist in various military institutions. It is shown that since 1925 until the end of life directed radiological office of the Central scientific research institute of tuberculosis M3 of the USSR. That circumstance that Prozorov along with 1926 for 1940 managed radiological office of hospital of S.P. Botkin is analyzed; in 1947-1949 headed department of a radiology of the 2nd MMI. A.E. Prozorov published more than 60 scientific works, including the monograph “Radiodiagnosis of a Pulmonary Tuberculosis” (1940), long time serving as one of the main guides to this question. A.E. Prozorov showed that the essential value in development of tubercular process has aggravation of the calmed-down tuberculous focuses (rein-fekt). It is noted that he paid attention to possibility of an atelectasis at tubercular bronchadenites. Its researches in the field of studying of the normal pulmonary drawing and roots of lungs and also descriptions of versions of the pulmonary drawing played a significant role for specification of early radiological manifestations of tuberculosis and other pulmonary diseases. A.E. Prozorov offered a method of the sliding mark for specification of localization of pleurisy, paid attention to a phenomenon of shift and change of a form of a shadow of an exudate at pleurisy in connection with breath phases. A number of works devoted to nonspecific diseases of lungs.

Patch-based lung ventilation estimation using multi-layer supervoxels

Patch-based approaches have received substantial attention over the recent years in medical imaging. One of their potential applications may be to provide more anatomically consistent ventilation maps estimated on dynamic lung CT. An assessment of regional lung function may act as a guide for radiotherapy, ensuring a more accurate treatment plan. This in turn, could spare well-functioning parts of the lungs. We present a novel method for lung ventilation estimation from dynamic lung CT imaging, combining a supervoxel-based image representation with deformations estimated during deformable image registration, performed between peak breathing phases. For this we propose a method that tracks changes of the intensity of previously extracted supervoxels. For the evaluation of the method we calculate correlation of the estimated ventilation maps with static ventilation images acquired from hyperpolarized Xenon129 MRI. We also investigate the influence of different image registration methods used to estimate deformations between the peak breathing phases in the dynamic CT imaging. We show that our method performs favorably to other ventilation estimation methods commonly used in the field, independently of the image registration method applied to dynamic CT. Due to the patch-based approach of our method, it may be physiologically more consistent with lung anatomy than previous methods relying on voxel-wise relationships. In our method the ventilation is estimated for supervoxels, which tend to group spatially close voxels with similar intensity values. The proposed method was evaluated on a dataset consisting of three lung cancer patients undergoing radiotherapy treatment, and this resulted in a correlation of 0.485 with XeMRI ventilation images, compared with 0.393 for the intensity-based approach, 0.231 for the Jacobian-based method and 0.386 for the Hounsfield units averaging method, on average. Within the limitation of the small number of cases analyzed, results suggest that the presented technique may be advantageous for CT-based ventilation estimation. The results showing higher values of correlation of the proposed method demonstrate the potential of our method to more accurately mimic the lung physiology.

Substance P Reshapes Inspiratory and Postinspiratory Activity Generated in Horizontal Brainstem Slices

The triple oscillator hypothesis posits that breathing is generated by three rhythmogenic microcircuits contained within the ventrolateral medulla. The preBötzinger Complex (preBötC) generates the inexorable inspiratory phase of breathing, whereas the conditional postinspiratory and active expiratory phases of breathing are generated by the Postinspiratory Complex (PiCo) and lateral parafacial respiratory group (PFL), respectively. Neuromodulators including norepinephrine, opioids, and somatostatin are known to differentially modulate the activity of each rhythmic circuit, likely contributing to the independence of each breathing phase and the flexibility of the respiratory rhythm. However, whether the functional boundaries of each of these microcircuits are static or can change dynamically based on the neuromodulatory state of the network is unknown. The neuromodulator Substance P (SP) facilitates breathing frequency in vivo and also in medullary slices that isolate the preBötC in vitro. Using a neonatal mouse horizontal brainstem slice preparation that preserves the rostrocaudal extent of the ventral medulla, we tested the hypothesis that SP differentially modulates bursting activity in the preBötC and PiCo and explored how SP may reshape the distribution of inspiratory activity within the wider medullary network. Rhythmic bursting activity was simultaneously recorded from the preBötC and PiCo (n=5). As expected, recordings from the PiCo, located ~500μm rostral to the preBötC, revealed a slow postinspiratory rhythm with robust bursts occurring after only ~1/12 preBötC bursts under control conditions. Following introduction of 0.5–1.0μM SP into the bath superfusion solution, PiCo bursts decreased in amplitude (~40%) and accelerated such that they occurred 1:1 with preBötC bursts without changing the time interval between peak preBötC and PiCo activity. In some PiCo recordings, we noted a small amount of inspiratory activity synchronized with preBötC bursts that was eliminated in the presence of SP. To test whether SP alters the spatial distribution of inspiratory activity along the ventral medulla, we simultaneously recorded inspiratory population activity generated at the preBötC and 400–600μm rostral (n=8). Following application of 1μM SP frequency was increased (~90%) and inspiratory burst amplitude was reduced by ~20% at the preBötC, whereas burst amplitude decreased by ~50% in rostral recordings. Based on these preliminary experiments, we hypothesize that SP reshapes breathing activity generated within the medulla by facilitating postinspiratory activity and reducing the extent of inspiratory activity rostral of the preBötC to shrink the size of the active inspiratory network. Differential neuromodulatory properties and dynamic functional boundaries of the rhythmogenic circuits that generate breathing may contribute to the remarkable flexibility and robustness of this vital physiological behavior.Support or Funding InformationNIH F32 HL134207NIH R01 HL126523This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.

Comparing the effectiveness and efficiency of various gating approaches for PBS proton therapy of pancreatic cancer using 4D-MRI datasets

Abdominal organ motion may lead to considerable uncertainties in pencil-beam scanning (PBS) proton therapy of pancreatic cancer. Beam gating, where irradiation only occurs in certain breathing phases in which the gating conditions are fulfilled, may be an option to reduce the interplay effect between tumor motion and the scanning beam. This study aims to, first, determine suitable gating windows with respect to effectiveness (low interplay effect) and efficiency (high duty cycles). Second, it investigates whether beam gating allows for a better mitigation of the interplay effect along the treatment course than free-breathing irradiations.Based on synthetic 4D-CTs, generated by warping 3D-CTs with vector fields extracted from time-resolved magnetic resonance imaging (4D-MRI) for 8 pancreatic cancer patients, 4D dose calculations (4DDC) were performed to analyze the duty cycle and homogeneity index HI = d5/d95 for four different gating scenarios. These were based on either fixed threshold values of CTV (clinical target volume) mean or maximum motion amplitudes (5 mm), relative CTV motion amplitudes (30%) or CTV overlap criteria (95%), respectively. 4DDC for 28-fractions treatment courses were performed with fixed and variable initial breathing phases to investigate the fractionation-induced mitigation of the interplay effect.Gating criteria, based on patient-specific relative 30% CTV motion amplitudes, showed the significantly best HI values with sufficient duty cycles, in contrast to inferior results by either fixed gating thresholds or overlap criteria. For gated treatments with 28 fractions, less fractionation-induced mitigation of the interplay effect was observed for gating criteria with gating windows ⩾30%, compared to free-breathing treatments. The gating effectiveness for multiple fractions was improved by allowing a variable initial breathing phase.Gating with relative amplitude thresholds are effective for proton therapy of pancreatic cancer. By combining beam gating with variable initial breathing phases, a pronounced mitigation of the interplay effect by fractionation can be achieved.

Microinjection of Gabazine and Kynurenic Acid into the PreBotzinger Complex Alter Cough Phase Durations in the Cat

The preBotzinger complex (preBotC) is important in breathing rhythmogenesis, but its role in cough patterning is unknown. We previously microinjected an excitatory neurotransmitter, 5 mM DL‐Homocysteic acid (DLH), into the preBotC, which altered breathing phase timing, but not cough phase timing. To further understand the role of the preBotC in cough rhythmogenesis, we hypothesized that inhibiting the preBotC would also have no effect on cough phase timing but would inhibit coughing. To test this hypothesis, we bilaterally microinjected 50 mM kynurenic acid (KYN, n=6), a broad spectrum excitatory amino acid blocker, and separately 0.1 mM gabazine (GBZ, n=5), a GABA‐A receptor antagonist, into the preBotC in anesthetized, spontaneously breathing adult cats. We recorded electromyograms (EMG) from parasternal (PS), diaphragm (DIA) and internal oblique (ABD) muscles. Cough was elicited by mechanical stimulation of the tracheobronchial tree. KYN microinjection significantly increased breathing frequency (21±2 to 43±10 breaths/min, p&lt;0.05) by changing both inspiratory (TI) and expiratory (TE) durations (TI 1.0±0.1 to 0.6±0.1, p&lt;0.05; TE 1.8±0.3 to 0.9±0.2 s, p&lt;0.05). DIA EMG magnitude and cough number did not change (4.5±0.6 to 3.9±1.4). However, inspiratory duration decreased (TI 1.1±0.1 to 0.7±0.06 s, p&lt;0.05), and while active expiratory duration (TE1) did not change significantly (TE1 1.0±0.11 to 0.8±0.04 s, n.s.), passive expiratory duration (TE2) was altered but not until 2 hours after injection (TE2 baseline: 0.9±0.2; 20 min after injection: 0.6±0.3 s; 2 hrs after injection: 0.50±0.2 s, p&lt;0.05). Both DIA (55±5 % of baseline) and ABD EMG (56±11 % of baseline) amplitudes decreased during cough. GBZ microinjection had the opposite effect on breathing compared to KYN, where breathing frequency decreased (21.0±4.0 to 12.0±1.0 breaths/minute, p&lt;0.05) by altering both TI and TE (TI 1.1±0.1 to 2.1 ±0.3 s, p&lt;0.05; TE 1.9±0.4 to 2.7±0.4 s, p&lt;0.05). With respect to cough, GBZ decreased cough number (5.1±0.6 to 1.5±0.4, p&lt;0.05) and ABD EMG magnitude (38±4 % of baseline, p&lt;0.05), although PS EMG magnitude did not change significantly (147±34 % of baseline). Microinjection of GBZ increased the duration of all phases of cough (TI 1.1±0.1 to 2.1±0.4 s, p&lt;0.05, TE1 0.9±0.05 to 1.8±0.2 s, p&lt;0.05, TE2 1.1±0.70 to 2.4±0.9 s, p=0.06). These data indicate that the preBotC participates in the regulation of cough phase timing and magnitudes in the anesthetized cat. Disruption of excitatory amino acid neurotransmission with KYN primarily affected inspiratory duration whereas GBZ‐induced blockade of GABA‐A receptor transmission had a generalized effect on cough. These data support an important but more complex role for the preBotC in the regulation of the cough motor pattern than that for breathing.Support or Funding InformationSupported by OT2OD023854‐01, HL131716.This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.

4-Dimensional computed tomography-based ventilation and compliance images for quantification of radiation-induced changes in pulmonary function.

4-Dimensional computed tomography (4DCT)-based ventilation imaging is a promising technique for evaluating pulmonary function, but lung elasticity and mechanics are usually not part of the ventilation image analysis. In this study we demonstrate a 4DCT-based imaging technique that can be used to calculate regional lung compliance changes after radiotherapy (RT). Six lung cancer patients were included in this study. Four of the patients had 4DCT images acquired pre-RT, 3 and 9months post-RT. Ventilation and compliance were calculated from the deformable image registration (DIR) of 4DCTs, performed from the end-inhale to the end-exhale breathing phase. Regional compliance was defined as the ratio of volumetric variation and associated stress in each voxel, representing lung elasticity and computed using a FEM-based framework. Ventilation, compliance and CT density were calculated for all pre-RT and post-RT 4DCTs and evaluation metrics were computed. Average CT density changes were 13.6±11.4HU after 3months and 26.9±15.8HU after 9months. Ventilation was reduced at 3months, but improved at 9months in regions with dose≥35Gy, encompassing about 10% of the lung volume; compliance was reduced at both time-points. Radiation dose≥35Gy caused major change in lung density and ventilation, which was higher than that previously reported in the literature (i.e. 24Gy). Lung tissue response is diverse with respect to CTdensity, ventilation and compliance. Combination of ventilation and compliance with CT density couldbe beneficial for understanding radiation-induced lung damage and consequently could help develop improved treatment protocols for lung cancer patients.

A spatially dynamic network underlies the generation of inspiratory behaviors.

The ability of neuronal networks to reconfigure is a key property underlying behavioral flexibility. Networks with recurrent topology are particularly prone to reconfiguration through changes in synaptic and intrinsic properties. Here, we explore spatial reconfiguration in the reticular networks of the medulla that generate breathing. Combined results from in vitro and in vivo approaches demonstrate that the network architecture underlying generation of the inspiratory phase of breathing is not static but can be spatially redistributed by shifts in the balance of excitatory and inhibitory network influences. These shifts in excitation/inhibition allow the size of the active network to expand and contract along a rostrocaudal medullary column during behavioral or metabolic challenges to breathing, such as changes in sensory feedback, sighing, and gasping. We postulate that the ability of this rhythm-generating network to spatially reconfigure contributes to the remarkable robustness and flexibility of breathing.

Conditional random fields for phase-based lung feature tracking with ultra-low-dose x-rays.

During radiation therapy, a continuous internal tumor monitoring without additional imaging dose is desirable. In this study, a sequential feature-based position estimation with ultra-low-dose (ULD) kV x rays using linear-chain conditional random fields (CRFs) is performed. Four-dimensional computed tomography (4D-CTs) of eight patients serve as a-priori information from which ULD projections are simulated using a Monte Carlo method. CRFs are trained with Local Energy-based Shape Histogram features extracted from the ULD images to estimate one out of ten breathing phases from the 4D-CT associated with the tumor position. Compared to a mean accuracy for ±1 breathing phase of 0.867 using a support vector machine (SVM), a mean accuracy of 0.958 results for the CRF with ten incident photons per pixel. This corresponds to a position estimation with a discretization error of 2.4-5.3mm assuming a linear displacement relation between the breathing phases and a systematic error of 2.0-4.4mm due to motion underestimation of the 4D-CT. The tumor position estimation is comparable to state-of-the-art methods despite its low imaging dose. Training CRFs further allows a prediction of the following phase and offers a precise post-treatment evaluation tool when decoding the full image sequence.

Automatic diaphragm segmentation for real-time lung tumor tracking on cone-beam CT projections: a convolutional neural network approach

Purpose: To automatically segment the diaphragm on individual lung cone-beam CT projection images, to enable real-time tracking of lung tumors using kilovoltage imaging. Methods: The deep neural network Mask R-CNN was trained on 3500 raw cone-beam CT projection images from 10 lung cancer patients, with the diaphragm manually segmented on each image used as a ground truth label. Ground-truth breathing traces were extracted from each patient for both diaphragm hemispheres, and apex positions were compared against the predicted output of the neural network. Ten-fold cross-validation was used to evaluate the segmentation accuracy. Results: The mean diaphragm apex prediction error was 4.4 mm. The mean percentage of projection images for which a successful prediction could me made was 87.3%. Prediction accuracy at some lateral gantry angles was worse due to overlap between diaphragm hemispheres, and the increased amount of fatty tissue. Conclusions: The neural network was able to track the diaphragm apex position successfully. This allows accurate assessment of the breathing phase, which can be used to estimate the position of the lung tumor in real time.

Analysis of wheeze sounds during tidal breathing according to severity levels in asthma patients

Objective: This study aimed to statistically analyze the behavior of time-frequency features in digital recordings of wheeze sounds obtained from patients with various levels of asthma severity (mild, moderate, and severe), and this analysis was based on the auscultation location and/or breath phase. Method: Segmented and validated wheeze sounds were collected from the trachea and lower lung base (LLB) of 55 asthmatic patients during tidal breathing maneuvers and grouped into nine different datasets. The quartile frequencies F25, F50, F75, F90 and F99, mean frequency (MF) and average power (AP) were computed as features, and a univariate statistical analysis was then performed to analyze the behavior of the time-frequency features. Results: All features generally showed statistical significance in most of the datasets for all severity levels [χ2 = 6.021–71.65, p < 0.05, η2 = 0.01–0.52]. Of the seven investigated features, only AP showed statistical significance in all the datasets. F25, F75, F90 and F99 exhibited statistical significance in at least six datasets [χ2 = 4.852–65.63, p < 0.05, η2 = 0.01–0.52], and F25, F50 and MF showed statistical significance with a large η2 in all trachea-related datasets [χ2 = 13.54–55.32, p < 0.05, η2 = 0.13–0.33]. Conclusion: The results obtained for the time-frequency features revealed that (1) the asthma severity levels of patients can be identified through a set of selected features with tidal breathing, (2) tracheal wheeze sounds are more sensitive and specific predictors of severity levels and (3) inspiratory and expiratory wheeze sounds are almost equally informative.

Investigation of thoracic four-dimensional CT-based dimension reduction technique for extracting the robust radiomic features

Robust feature selection in radiomic analysis is often implemented using the RIDER test-retest datasets. However, the CT Protocol between the facility and test-retest datasets are different. Therefore, we investigated possibility to select robust features using thoracic four-dimensional CT (4D-CT) scans that are available from patients receiving radiation therapy.In 4D-CT datasets of 14 lung cancer patients who underwent stereotactic body radiotherapy (SBRT) and 14 test-retest datasets of non-small cell lung cancer (NSCLC), 1170 radiomic features (shape: n = 16, statistics: n = 32, texture: n = 1122) were extracted. A concordance correlation coefficient (CCC) > 0.85 was used to select robust features. We compared the robust features in various 4D-CT group with those in test-retest.The total number of robust features was a range between 846/1170 (72%) and 970/1170 (83%) in all 4D-CT groups with three breathing phases (40%–60%); however, that was a range between 44/1170 (4%) and 476/1170 (41%) in all 4D-CT groups with 10 breathing phases. In test-retest, the total number of robust features was 967/1170 (83%); thus, the number of robust features in 4D-CT was almost equal to that in test-retest by using 40–60% breathing phases.In 4D-CT, respiratory motion is a factor that greatly affects the robustness of features, thus by using only 40–60% breathing phases, excessive dimension reduction will be able to be prevented in any 4D-CT datasets, and select robust features suitable for CT protocol of your own facility.