Objective.There is an increasing interest in calculating and measuring linear energy transfer (LET) spectra in particle therapy in order to assess their impact in biological terms. As such, the accuracy of the particle fluence energy spectra becomes paramount. This study focuses on quantifying energy depositions of distinct proton, helium, carbon, and oxygen ion beams using a silicon pixel detector developed at CERN to determine LET spectra in silicon.Approach.While detection systems have been investigated in this pursuit, the scarcity of detectors capable of providing per-ion data with high spatial and temporal resolution remains an issue. This gap is where silicon pixel detector technology steps in, enabling online tracking of single-ion energy deposition. The used detector consisted of a 300μm thick silicon sensor operated in partial depletion.Main results.During post-processing, artifacts in the acquired signals were identified and methods for their corrections were developed. Subsequently, a correlation between measured and Monte Carlo-based simulated energy deposition distributions was performed, relying on a two-step recalibration approach based on linear and saturating exponential models. Despite the observed saturation effects, deviations were confined below 7% across the entire investigated range of track-averaged LET values in silicon from 0.77 keVμm-1to 93.16 keVμm-1.Significance.Simulated and measured mean energy depositions were found to be aligned within 7%, after applying artifact corrections. This extends the range of accessible LET spectra in silicon to clinically relevant values and validates the accuracy and reliability of the measurements. These findings pave the way towards LET-based dosimetry through an approach to translate these measurements to LET spectra in water. This will be addressed in a future study, extending functionality of treatment planning systems into clinical routine, with the potential of providing ion-beam therapy of utmost precision to cancer patients.
- MeSH
- Silicon MeSH
- Linear Energy Transfer * MeSH
- Monte Carlo Method MeSH
- Radiometry instrumentation MeSH
- Publication type
- Journal Article MeSH
OBJECTIVES: To assess the feasibility of acquiring adequate transperineal ultrasound (TPUS) volumes of the anal sphincter (AS) immediately after vaginal birth, the reproducibility of its measurements, and detecting defects therein. METHODS: Secondary analysis of TPUS volumes of the AS, acquired immediately after vaginal birth with a transversely oriented convex probe. Two independent experts ranked off-line image quality as "inadequate," "adequate," or "ideal" using the Point-of-Care Ultrasound Image Quality scale. On "adequate" and "ideal" quality volumes, the length of the external AS at 6 and 12 o'clock, and the volume of the external and internal AS were measured. Additionally, volumes were screened for AS defects on tomographic ultrasound imaging. Subsequently, we rated the intra- and interrater agreement on those findings. RESULTS: Of 183 volumes, 162 were considered "adequate" or of "ideal" quality (88.5%). Reasons for "inadequacy" were shadow artifacts (16/21), poor resolution (3/21), incomplete acquisition (1/21), or aberrant AS morphology (1/21). The intrarater reliability of two-dimensional (2D) and three-dimensional (3D) measurements was excellent, whereas interrater reliability was fair to good for 2D measurements and good for 3D measurements. In those tomographic ultrasound imaging (TUI) sequences including AS defects, the intra- and interrater reliability of the defect measurement were excellent [intraclass correlation coefficient (ICC) = 0.92 (0.80-0.94)] and moderate [ICC = 0.72 (0.63-0.79)]. In this cohort, there were only few (4/48; 8.3%) AS defects. However, grading them was poorly reproducible between experts. CONCLUSION: TPUS of the AS immediately after vaginal birth yields adequate image quality and allows for reproducible measurements. In the few patients with AS defects, there was good agreement on the presence, but it was poor for the extent of defects.
- MeSH
- Anal Canal * diagnostic imaging MeSH
- Adult MeSH
- Humans MeSH
- Perineum diagnostic imaging MeSH
- Reproducibility of Results MeSH
- Feasibility Studies * MeSH
- Pregnancy MeSH
- Ultrasonography * methods MeSH
- Imaging, Three-Dimensional * methods MeSH
- Check Tag
- Adult MeSH
- Humans MeSH
- Pregnancy MeSH
- Female MeSH
- Publication type
- Journal Article MeSH
High-frequency waveform recordings of biological signals enable more detailed data analysis and deeper physiological exploration. However, the waveform data—like invasive arterial blood pressure (ABP)—are particularly susceptible to frequent contamination with artifacts that can devalue the subsequent calculations like pressure reactivity index (PRx). This study aimed to verify the ability of the short-time Fourier transform (STFT) based algorithm to detect artifacts in the ABP waveform. Four types of modeled artifacts (rectangular, fast impulse, sawtooth and baseline drift) with different durations and amplitudes were inserted into undisturbed ABP waveforms. Short-time Fourier transform with a 5-second time window is computed on artifact-polluted ABP signals to detect changes in the frequency domain caused by these artifacts. An algorithm with three decision-making rules based on the dominant frequency component, standardized power spectrum, and the value of the second harmonic of the dominant frequency was used. Only segments that passed all three rules were labeled as artifact-free. Results indicated high sensitivity (93.35% and 94.83%) in detecting rectangular and sawtooth artifacts, with specificity exceeding 99% for both. Baseline drift artifact was detected with a low sensitivity of 5.02%, and fast impulse was not detected. This study proposes the application of a short-time Fourier transform-based algorithm to enhance the detection of clinically significant artifacts in arterial blood pressure signals, particularly relevant for PRx and other secondary calculations.
OBJECTIVES: Accurate detection of metastatic brain lesions (MBL) is critical due to advances in radiosurgery. We compared the results of three readers in detecting MBL using T1-weighted 2D spin echo (SE) and sampling perfection with application-optimized contrasts using different flip angle evolution (SPACE) sequences with whole-brain coverage at both 1.5 T and 3 T. METHODS: Fifty-six patients evaluated for MBL were included and underwent a standard protocol (1.5 T, n = 37; 3 T, n = 19), including postcontrast T1-weighted SE and SPACE. The rating was performed by three raters in two sessions > six weeks apart. The true number of MBL was determined using all available imaging including follow-up. Intraclass correlations for intra-rater and inter-rater agreement were calculated. Signal intensity ratios (SIR; enhancing lesion, white matter) were determined on a subset of 46 MBL > 4 mm. A paired t-test was used to evaluate postcontrast sequence order and SIR. Reader accuracy was evaluated by the coefficient of determination. RESULTS: A total of 135 MBL were identified (mean/subject 2.41, SD 6.4). The intra-rater agreement was excellent for all 3 raters (ICC = 0.97-0.992), as was the inter-rater agreement (ICC = 0.995 SE, 0.99 SPACE). Subjective qualitative ratings were lower for SE images; however, signal intensity ratios were higher in SE sequences. Accuracy was high in all readers for both SE (R2 0.95-0.96) and SPACE (R2 0.91-0.96) sequences. CONCLUSIONS: Although SE sequences are superior to gradient echo sequences in the detection of small MBL, they have long acquisition times and frequent artifacts. We show that T1-weighted SPACE is not inferior to standard thin-slice SE sequences in the detection of MBL at both imaging fields. CRITICAL RELEVANCE STATEMENT: Our results show the suitability of 3D T1-weighted turbo spin echo (TSE) sequences (SPACE, CUBE, VISTA) in the detection of brain metastases at both 1.5 T and 3 T. KEY POINTS: • Accurate detection of brain metastases is critical due to advances in radiosurgery. • T1-weighted SE sequences are superior to gradient echo in detecting small metastases. • T1-weighted 3D-TSE sequences may achieve high resolution and relative insensitivity to artifacts. • T1-weighted 3D-TSE sequences have been recommended in imaging brain metastases at 3 T. • We found T1-weighted 3D-TSE equivalent to thin-slice SE at 1.5 T and 3 T.
- Publication type
- Journal Article MeSH
OBJECTIVE: Recent studies have integrated copy number variant (CNV) and gene analysis using target enrichment. Here, we transferred this concept to our routine genetics laboratory, which is not linked to centralized non-invasive prenatal testing (NIPT) facilities. METHOD: From a cohort of 100 pregnant women, 22 were selected for the analysis of maternal genomic DNA (gDNA) along with fetal cell-free DNA. Using targeted enrichment, 135 genes were analyzed, combined with aberrations of chromosomes 21, 18, 13, X, and Y. The data were subjected to specificity and sensitivity analyses, and correlated with the results from invasive testing methods. RESULTS: The sensitivity/specificity was determined for the CNV analysis of chromosomes: 21 (80%/75%), 18 (-/82%), 13 (100%/67%), and Y (100%/100%). The gene detection was valid for maternal gDNA. However, for cell-free fetal DNA, it was not possible to determine the boundary between an artifact and a real sequence variant. CONCLUSION: The target enrichment method combining CNV and gene detection seems feasible in a regular laboratory. However, this method can only be responsibly optimized with a sufficient number of controls and further validation on a strong bioinformatic background. The present results showed that NIPT should be performed in specialized centers, and that its introduction to isolated laboratories may not provide valid data.
Manual visual review, annotation and categorization of electroencephalography (EEG) is a time-consuming task that is often associated with human bias and requires trained electrophysiology experts with specific domain knowledge. This challenge is now compounded by development of measurement technologies and devices allowing large-scale heterogeneous, multi-channel recordings spanning multiple brain regions over days, weeks. Currently, supervised deep-learning techniques were shown to be an effective tool for analyzing big data sets, including EEG. However, the most significant caveat in training the supervised deep-learning models in a clinical research setting is the lack of adequate gold-standard annotations created by electrophysiology experts. Here, we propose a semi-supervised machine learning technique that utilizes deep-learning methods with a minimal amount of gold-standard labels. The method utilizes a temporal autoencoder for dimensionality reduction and a small number of the expert-provided gold-standard labels used for kernel density estimating (KDE) maps. We used data from electrophysiological intracranial EEG (iEEG) recordings acquired in two hospitals with different recording systems across 39 patients to validate the method. The method achieved iEEG classification (Pathologic vs. Normal vs. Artifacts) results with an area under the receiver operating characteristic (AUROC) scores of 0.862 ± 0.037, 0.879 ± 0.042, and area under the precision-recall curve (AUPRC) scores of 0.740 ± 0.740, 0.714 ± 0.042. This demonstrates that semi-supervised methods can provide acceptable results while requiring only 100 gold-standard data samples in each classification category. Subsequently, we deployed the technique to 12 novel patients in a pseudo-prospective framework for detecting Interictal epileptiform discharges (IEDs). We show that the proposed temporal autoencoder was able to generalize to novel patients while achieving AUROC of 0.877 ± 0.067 and AUPRC of 0.705 ± 0.154.
Úvod: Hrudní pás umožňuje pořídit 1svodový EKG záznam. Získaná data byla validována pro měření srdeční frekvence a rovněž i pro detekci fibrilace síní díky srovnání s krátkými EKG záznamy z holterovského EKG měření u selektovaných pacientů. Zatím ale nebyla ověřena možnost vyhodnocení dlouhých EKG záznamů u neselektovaných kardiologických pacientů se širokým spektrem srdečních chorob. Metodologie a výsledky: Do studie bylo zařazeno 54 hospitalizovaných a 53 ambulantních pacientů a 54 zdravých kontrol (n = 161 celkově). U všech účastníků studie byl pomocí hrudního pásu Polar H10 pořízen 1-2hodinový EKG záznam (celkově 1 153 229 úderů srdce; průměrná srdeční frekvence 76,6/min; sinusový rytmus u 86,3 %, fibrilace síní zjištěna u 13,7 %; dokumentováno 0,46 % síňových extrasystol a 0,49 % komorových extrasystol). Z výše uvedeného počtu 1 153 229 srdečních tepů jich 1 128 319 bylo hodnoceno lékařem jako snadno interpretovatelných. Celkově tak bylo 2,16 % záznamu vyhodnoceno jako obtížně interpretovatelný nebo neinterpretovatelný šum (A: 2,31 %; B: 1,95 %; C: 2,20 %). Z EKG záznamu z hrudního pásu lékař při srovnání s 12svodovým EKG záznamem spolehlivě určil základní srdeční rytmus u většiny účastníků (u 51/54 [94,4 %] hospitalizovaných pacientů a u 100 % ambulantních pacientů a zdravých kontrol). U tří jedinců byl základní rytmus na EKG vyhodnocen jako nejasný. U všech tří byly všechny komplexy QRS stimulované. U hospitalizovaných pacientů byl EKG záznam z hrudního pásu zobrazený v reálném čase na mobilním telefonu srovnatelný s EKG záznamem z telemetrického monitorování (shoda v 53 z 54 případů; 98,1 %). Závěr: EKG záznam z hrudního pásu, pořízený u hospitalizovaných i ambulantních pacientů s různými typy poruch srdečního rytmu, stejně tak jako u zdravých kontrol, lze v každodenní praxi použít pro zhodnocení základního srdečního rytmu, záchyt fibrilace síní i extrasystol, a to při minimálním procentu obtížně hodnotitelných záznamů. Opatrnosti je třeba při interpretaci EKG záznamu u pacientů se stimulovaným rytmem a u pacientů s flutterem síní. Hrudní pás je tak možno použít pro kontinuální EKG monitorování, hodnocení srdečního rytmu i screening fibrilace síní.
Background: The chest-belt can be used to obtain a 1-lead ECG. Data from it have been validated for the determination of heart rate and for the possibility to detect atrial fibrillation (AF) compared to ECG-Holter on a short ECG recording in selected patients. However, validation of the possibility to evaluate long ECG recordings in patients with a wide range of heart diseases has not yet been performed. Methodology and results: 54 hospitalized patients, 53 outpatients and 54 healthy controls were enrolled in the study (n = 161 in total). Using a Polar H10 chest-belt, 1-2 hours of ECG were recorded in all patients (1 153 229 heartbeats, average heart rate 76.6/min, 86.3% in sinus rhythm, 13.7% with atrial fibrillation, 0.46% atrial premature beats, 0.49% ventricular premature beats). The presence of noise was 2.16% (A: 2.31%; B: 1.95%; C: 2.20%). 1 128 319 /1 153 229 were evaluated as easy to interpret. Using ECG from the belt, the basic rhythm was reliably determined by the physician in majority of patients (51/54, 94.4% in hospitalized patients; in 100% of outpatients and healthy controls) when compared to 12-lead ECG. 3 cases were evaluated as unclear; in all of these cases, all QRS complexes were stimulated by a pacemaker. In hospitalized patients, real-time ECG from the belt was comparable to telemetric ECG monitoring (match in 53/54, 98.1%). Conclusion: The ECG obtained from the chest-belt in hospitalized patients and outpatients with a wide range of cardiovascular diseases, as well as in healthy individuals, is usable in real practice for evaluation of baseline rhythm, atrial fibrillation and premature contractions with a minimal proportion of difficulties to interpret recordings due to artefacts. Caution should be exercised in interpretation of the ECG in patients with stimulated rhythm and in patients with atrial flutter. The chest belt can be used as a means for continuous monitoring of ECG, evaluation of rhythm and screening of atrial fibrillation.
Fetal phonocardiography is a non-invasive, completely passive and low-cost method based on sensing acoustic signals from the maternal abdomen. However, different types of interference are sensed along with the desired fetal phonocardiography. This study focuses on the comparison of fetal phonocardiography filtering using eight algorithms: Savitzky-Golay filter, finite impulse response filter, adaptive wavelet transform, maximal overlap discrete wavelet transform, variational mode decomposition, empirical mode decomposition, ensemble empirical mode decomposition, and complete ensemble empirical mode decomposition with adaptive noise. The effectiveness of those methods was tested on four types of interference (maternal sounds, movement artifacts, Gaussian noise, and ambient noise) and eleven combinations of these disturbances. The dataset was created using two synthetic records r01 and r02, where the record r02 was loaded with higher levels of interference than the record r01. The evaluation was performed using the objective parameters such as accuracy of the detection of S1 and S2 sounds, signal-to-noise ratio improvement, and mean error of heart interval measurement. According to all parameters, the best results were achieved using the complete ensemble empirical mode decomposition with adaptive noise method with average values of accuracy = 91.53% in the detection of S1 and accuracy = 68.89% in the detection of S2. The average value of signal-to-noise ratio improvement achieved by complete ensemble empirical mode decomposition with adaptive noise method was 9.75 dB and the average value of the mean error of heart interval measurement was 3.27 ms.
While various QRS detection and classification methods were developed in the past, the Holter ECG data acquired during daily activities by wearable devices represent new challenges such as increased noise and artefacts due to patient movements. Here, we present a deep-learning model to detect and classify QRS complexes in single-lead Holter ECG. We introduce a novel approach, delivering QRS detection and classification in one inference step. We used a private dataset (12,111 Holter ECG recordings, length of 30 s) for training, validation, and testing the method. Twelve public databases were used to further test method performance. We built a software tool to rapidly annotate QRS complexes in a private dataset, and we annotated 619,681 QRS complexes. The standardised and down-sampled ECG signal forms a 30-s long input for the deep-learning model. The model consists of five ResNet blocks and a gated recurrent unit layer. The model's output is a 30-s long 4-channel probability vector (no-QRS, normal QRS, premature ventricular contraction, premature atrial contraction). Output probabilities are post-processed to receive predicted QRS annotation marks. For the QRS detection task, the proposed method achieved the F1 score of 0.99 on the private test set. An overall mean F1 cross-database score through twelve external public databases was 0.96 ± 0.06. In terms of QRS classification, the presented method showed micro and macro F1 scores of 0.96 and 0.74 on the private test set, respectively. Cross-database results using four external public datasets showed micro and macro F1 scores of 0.95 ± 0.03 and 0.73 ± 0.06, respectively. Presented results showed that QRS detection and classification could be reliably computed in one inference step. The cross-database tests showed higher overall QRS detection performance than any of compared methods.
- MeSH
- Algorithms MeSH
- Artifacts MeSH
- Electrocardiography, Ambulatory methods MeSH
- Electrocardiography methods MeSH
- Ventricular Premature Complexes * MeSH
- Humans MeSH
- Wearable Electronic Devices * MeSH
- Signal Processing, Computer-Assisted MeSH
- Check Tag
- Humans MeSH
- Publication type
- Journal Article MeSH
- Research Support, Non-U.S. Gov't MeSH
With the rapid advancement of sequencing technologies, next generation sequencing (NGS) analysis has been widely applied in cancer genomics research. More recently, NGS has been adopted in clinical oncology to advance personalized medicine. Clinical applications of precision oncology require accurate tests that can distinguish tumor-specific mutations from artifacts introduced during NGS processes or data analysis. Therefore, there is an urgent need to develop best practices in cancer mutation detection using NGS and the need for standard reference data sets for systematically measuring accuracy and reproducibility across platforms and methods. Within the SEQC2 consortium context, we established paired tumor-normal reference samples and generated whole-genome (WGS) and whole-exome sequencing (WES) data using sixteen library protocols, seven sequencing platforms at six different centers. We systematically interrogated somatic mutations in the reference samples to identify factors affecting detection reproducibility and accuracy in cancer genomes. These large cross-platform/site WGS and WES datasets using well-characterized reference samples will represent a powerful resource for benchmarking NGS technologies, bioinformatics pipelines, and for the cancer genomics studies.
- MeSH
- Benchmarking MeSH
- Genome, Human * MeSH
- Genomics MeSH
- Precision Medicine MeSH
- Humans MeSH
- Cell Line, Tumor MeSH
- Neoplasms genetics MeSH
- Whole Genome Sequencing * MeSH
- Exome Sequencing * MeSH
- Computational Biology MeSH
- Check Tag
- Humans MeSH
- Publication type
- Journal Article MeSH
- Dataset MeSH
- Research Support, Non-U.S. Gov't MeSH
- Research Support, N.I.H., Extramural MeSH
