Search

Abstract Research on different machine learning (ML) has become incredibly popular during the past few decades. However, for some researchers not familiar with statistics, it might be difficult to understand how to evaluate the performance of ML models and compare them with each other. Here, we introduce the most common evaluation metrics used for the typical supervised ML tasks including binary, multi-class, and multi-label classification, regression, image segmentation, object detection, and information retrieval. We explain how to choose a suitable statistical test for comparing models, how to obtain enough values of the metric for testing, and how to perform the test and interpret its results. We also present a few practical examples about comparing convolutional neural networks used to classify X-rays with different lung infections and detect cancer tumors in positron emission tomography images.

Background: A generative adversarial network (GAN) has gained popularity as a data augmentation technique in the medical field due to its efficiency in creating synthetic data for different machine learning models. In particular, the earlier literature suggests that the classification accuracy of a convolutional neural network (CNN) used for detecting brain tumors in magnetic resonance imaging (MRI) images increases when GAN-generated images are included in the training data together with the original images. However, there is little research about how the exact number of GAN-generated images and their ratio to the original images affects the results obtained. Materials and methods: Here, by using 1000 original images from a public repository with MRI images of patients with or without brain tumors, we built a GAN model to create synthetic brain MRI images. A modified U-Net CNN is trained multiple times with different training datasets and its classification accuracy is evaluated from a separate test set of another 1000 images. The Mann–Whitney U test is used to estimate whether the differences in the accuracy caused by different choices of training data are statistically significant. Results: According to our results, the use of GAN augmentation only sometimes produces a significant improvement. For instance, the classification accuracy significantly increases when 250–750 GAN-generated images are added to 1000 original images (p-values ≤ 0.0025) but decreases when 10 GAN-generated images are added to 500 original images (p-value: 0.03). Conclusions: Whenever GAN-based augmentation is used, the number of GAN-generated images should be carefully considered while accounting for the number of original images.

This perspective paper explores the untapped potential of artificial intelligence (AI), particularly machine learning-based dimension reduction techniques in multimodal neuroimaging analysis of Long COVID fatigue. The complexity and high dimensionality of neuroimaging data from modalities such as positron emission tomography (PET) and magnetic resonance imaging (MRI) pose significant analytical challenges. Deep neural networks and other machine learning approaches offer powerful tools for managing this complexity and extracting meaningful patterns. The paper discusses current challenges in neuroimaging data analysis, reviews state-of-the-art AI approaches for dimension reduction and multimodal integration, and examines their potential applications in Long COVID research. Key areas of focus include the development of AI-based biomarkers, AI-informed treatment strategies, and personalized medicine approaches. The authors argue that AI-driven multimodal neuroimaging analysis represents a paradigm shift in studying complex brain disorders like Long COVID. While acknowledging technical and ethical challenges, the paper emphasizes the potential of these advanced techniques to uncover new insights into the condition, which might lead to improved diagnostic and therapeutic strategies for those affected by Long COVID fatigue. The broader implications for understanding and treating other complex neurological and psychiatric conditions are also discussed.

The stability–plasticity dilemma remains a critical challenge in developing artificial intelligence (AI) systems capable of continuous learning. This perspective paper presents a novel approach by drawing inspiration from the mammalian hippocampus–cortex system. We elucidate how this biological system’s ability to balance rapid learning with long-term memory retention can inspire novel AI architectures. Our analysis focuses on key mechanisms, including complementary learning systems and memory consolidation, with emphasis on recent discoveries about sharp-wave ripples and barrages of action potentials. We propose innovative AI designs incorporating dual learning rates, offline consolidation, and dynamic plasticity modulation. This interdisciplinary approach offers a framework for more adaptive AI systems while providing insights into biological learning. We present testable predictions and discuss potential implementations and implications of these biologically inspired principles. By bridging neuroscience and AI, our perspective aims to catalyze advancements in both fields, potentially revolutionizing AI capabilities while deepening our understanding of neural processes.

Abstract The aim of this study was to develop a convolutional neural network (CNN) for classifying positron emission tomography (PET) images of patients with and without head and neck squamous cell carcinoma (HNSCC) and other types of head and neck cancer. A PET/magnetic resonance imaging scan with 18F-fluorodeoxyglucose (18F-FDG) was performed for 200 head and neck cancer patients, 182 of which were diagnosed with HNSCC, and the location of cancer tumors was marked to the images with a binary mask by a medical doctor. The models were trained and tested with five-fold cross-validation with the primary data set of 1990 2D images obtained by dividing the original 3D images of 178 HNSCC patients into transaxial slices and with an additional test set with 238 images from the patients with head and neck cancer other than HNSCC. A shallow and a deep CNN were built by using the U-Net architecture for classifying the data into two groups based on whether an image contains cancer or not. The impact of data augmentation on the performance of the two CNNs was also considered. According to our results, the best model for this task in terms of area under receiver operator characteristic curve (AUC) is a deep augmented model with a median AUC of 85.1%. The four models had highest sensitivity for HNSCC tumors on the root of the tongue (median sensitivities of 83.3–97.7%), in fossa piriformis (80.2–93.3%), and in the oral cavity (70.4–81.7%). Despite the fact that the models were trained with only HNSCC data, they had also very good sensitivity for detecting follicular and papillary carcinoma of thyroid gland and mucoepidermoid carcinoma of the parotid gland (91.7–100%).

Long COVID (Coronavirus disease), affecting millions globally, presents unprecedented challenges to healthcare systems due to its complex, multifaceted nature and the lack of effective treatments. This perspective review explores the potential of artificial intelligence (AI)-guided transcranial direct current stimulation (tDCS) as an innovative approach to address the urgent need for effective Long COVID management. The authors examine how AI could optimize tDCS protocols, enhance clinical trial design, and facilitate personalized treatment for the heterogeneous manifestations of Long COVID. Key areas discussed include AI-driven personalization of tDCS parameters based on individual patient characteristics and real-time symptom fluctuations, the use of machine learning for patient stratification, and the development of more sensitive outcome measures in clinical trials. This perspective addresses ethical considerations surrounding data privacy, algorithmic bias, and equitable access to AI-enhanced treatments. It also explores challenges and opportunities for implementing AI-guided tDCS across diverse healthcare settings globally. Future research directions are outlined, including the need for large-scale validation studies and investigations of long-term efficacy and safety. The authors argue that while AI-guided tDCS shows promise for addressing the complex nature of Long COVID, significant technical, ethical, and practical challenges remain. They emphasize the importance of interdisciplinary collaboration, patient-centered approaches, and a commitment to global health equity in realizing the potential of this technology. This perspective article provides a roadmap for researchers, clinicians, and policymakers involved in developing and implementing AI-guided neuromodulation therapies for Long COVID and potentially other neurological and psychiatric conditions.

Background: The emergence of new SARS-CoV-2 variants with significant immune-evasiveness, the relaxation of measures for reducing the number of infections, the waning of immune protection (particularly in high-risk population groups), and the low uptake of new vaccine boosters, forecast new waves of hospitalizations and admission to intensive care units. There is an urgent need for easily implementable and clinically effective Early Warning Scores (EWSs) that can predict the risk of complications within the next 24–48 hr. Although EWSs have been used in the evaluation of COVID-19 patients, there are several clinical limitations to their use. Moreover, no models have been tested on geographically distinct populations or population groups with varying levels of immune protection. Methods: We developed and validated COVID-19 Early Warning Score (COEWS), an EWS that is automatically calculated solely from laboratory parameters that are widely available and affordable. We benchmarked COEWS against the widely used NEWS2. We also evaluated the predictive performance of vaccinated and unvaccinated patients. Results: The variables of the COEWS predictive model were selected based on their predictive coefficients and on the wide availability of these laboratory variables. The final model included complete blood count, blood glucose, and oxygen saturation features. To make COEWS more actionable in real clinical situations, we transformed the predictive coefficients of the COEWS model into individual scores for each selected feature. The global score serves as an easy-to-calculate measure indicating the risk of a patient developing the combined outcome of mechanical ventilation or death within the next 48 hr. Conclusions: The COEWS score predicts death or MV within the next 48 hr based on routine and widely available laboratory measurements. The extensive external validation, its high performance, its ease of use, and its positive benchmark in comparison with the widely used NEWS2 position COEWS as a new reference tool for assisting clinical decisions and improving patient care in the upcoming pandemic waves. Funding: University of Vienna.

Abstract Background Attenuation correction is crucial in quantitative positron emission tomography-magnetic resonance (PET-MRI) imaging. We evaluated three methods to improve the segmentation and modelling of the attenuation coefficients in the nasal sinus region. Two methods (cuboid and template method) included a MRI-CT conversion model for assigning the attenuation coefficients in the nasal sinus region, whereas one used fixed attenuation coefficient assignment (bulk method). Methods The study population consisted of data of 10 subjects which had undergone PET-CT and PET-MRI. PET images were reconstructed with and without time-of-flight (TOF) using CT-based attenuation correction (CTAC) as reference. Comparison was done visually, using DICE coefficients, correlation, analyzing attenuation coefficients, and quantitative analysis of PET and bias atlas images. Results The median DICE coefficients were 0.824, 0.853, 0.849 for the bulk, cuboid and template method, respectively. The median attenuation coefficients were 0.0841 cm−1, 0.0876 cm−1, 0.0861 cm−1 and 0.0852 cm−1, for CTAC, bulk, cuboid and template method, respectively. The cuboid and template methods showed error of less than 2.5% in attenuation coefficients. An increased correlation to CTAC was shown with the cuboid and template methods. In the regional analysis, improvement in at least 49% and 80% of VOI was seen with non-TOF and TOF imaging. All methods showed errors less than 2.5% in non-TOF and less than 2% in TOF reconstructions. Conclusions We evaluated two proof-of-concept methods for improving quantitative accuracy in PET/MRI imaging and showed that bias can be further reduced by inclusion of TOF. Largest improvements were seen in the regions of olfactory bulb, Heschl's gyri, lingual gyrus and cerebellar vermis. However, the overall effect of inclusion of the sinus region as separate class in MRAC to PET quantification in the brain was considered modest.

Background and purpose: Periprosthetic joint infection (PJI) is the commonest reason for revision after total knee arthroplasty (TKA). We assessed the risk factors for revision due to PJI following TKA based on the Finnish Arthroplasty Register (FAR). Patients and methods: We analyzed 62,087 primary condylar TKAs registered between June 2014 and February 2020 with revision for PJI as the endpoint. Cox proportional hazards regression was used to estimate hazard ratios (HR) with 95% confidence intervals (CI) for the first PJI revision using 25 potential patient- and surgical-related risk factors as covariates. Results: 484 knees were revised for the first time during the first postoperative year because of PJI. The HRs for revision due to PJI in unadjusted analysis were 0.5 (0.4–0.6) for female sex, 0.7 (0.6–1.0) for BMI 25–29, and 1.6 (1.1–2.5) for BMI > 40 compared with BMI < 25, 4.0 (1.3–12) for preoperative fracture diagnosis compared with osteoarthritis, and 0.7 (0.5–0.9) for use of an antimicrobial incise drape. In adjusted analysis the HRs were 2.2 (1.4–3.5) for ASA class III–IV compared with class I, 1.7 (1.4–2.1) for intraoperative bleeding ≥ 100 mL, 1.4 (1.2–1.8) for use of a drain, 0.7 (0.5–1.0) for short duration of operation of 45–59 minutes, and 1.7 (1.3–2.3) for long operation duration > 120 min compared with 60–89 minutes, and 1.3 (1.0–1.8) for use of general anesthesia. Conclusion: We found increased risk for revision due to PJI when no incise drape was used. The use of drainage also increased the risk. Specializing in performing TKA reduces operative time and thereby also the PJI rate.

Abstract We implemented a two-dimensional convolutional neural network (CNN) for classification of polar maps extracted from Carimas (Turku PET Centre, Finland) software used for myocardial perfusion analysis. 138 polar maps from 15O–H2O stress perfusion study in JPEG format from patients classified as ischemic or non-ischemic based on finding obstructive coronary artery disease (CAD) on invasive coronary artery angiography were used. The CNN was evaluated against the clinical interpretation. The classification accuracy was evaluated with: accuracy (ACC), area under the receiver operating characteristic curve (AUC), F1 score (F1S), sensitivity (SEN), specificity (SPE) and precision (PRE). The CNN had a median ACC of 0.8261, AUC of 0.8058, F1S of 0.7647, SEN of 0.6500, SPE of 0.9615 and PRE of 0.9286. In comparison, clinical interpretation had ACC of 0.8696, AUC of 0.8558, F1S of 0.8333, SEN of 0.7500, SPE of 0.9615 and PRE of 0.9375. The CNN classified only 2 cases differently than the clinical interpretation. The clinical interpretation and CNN had similar accuracy in classifying false positives and true negatives. Classification of ischemia is feasible in 15O–H2O stress perfusion imaging using JPEG polar maps alone with a custom CNN and may be useful for the detection of obstructive CAD.

Clustering time activity curves of PET images have been used to separate clinically relevant areas of the brain or tumours. However, PET image segmentation in multiorgan level is much less studied due to the available total-body data being limited to animal studies. Now, the new PET scanners providing the opportunity to acquire total-body PET scans also from humans are becoming more common, which opens plenty of new clinically interesting opportunities. Therefore, organ-level segmentation of PET images has important applications, yet it lacks sufficient research. In this proof of concept study, we evaluate if the previously used segmentation approaches are suitable for segmenting dynamic human total-body PET images in organ level. Our focus is on general-purpose unsupervised methods that are independent of external data and can be used for all tracers, organisms, and health conditions. Additional anatomical image modalities, such as CT or MRI, are not used, but the segmentation is done purely based on the dynamic PET images. The tested methods are commonly used building blocks of the more sophisticated methods rather than final methods as such, and our goal is to evaluate if these basic tools are suited for the arising human total-body PET image segmentation. First, we excluded methods that were computationally too demanding for the large datasets from human total-body PET scanners. These criteria filtered out most of the commonly used approaches, leaving only two clustering methods, k-means and Gaussian mixture model (GMM), for further analyses. We combined k-means with two different preprocessing approaches, namely, principal component analysis (PCA) and independent component analysis (ICA). Then, we selected a suitable number of clusters using 10 images. Finally, we tested how well the usable approaches segment the remaining PET images in organ level, highlight the best approaches together with their limitations, and discuss how further research could tackle the observed shortcomings. In this study, we utilised 40 total-body [18F] fluorodeoxyglucose PET images of rats to mimic the coming large human PET images and a few actual human total-body images to ensure that our conclusions from the rat data generalise to the human data. Our results show that ICA combined with k-means has weaker performance than the other two computationally usable approaches and that certain organs are easier to segment than others. While GMM performed sufficiently, it was by far the slowest one among the tested approaches, making k-means combined with PCA the most promising candidate for further development. However, even with the best methods, the mean Jaccard index was slightly below 0.5 for the easiest tested organ and below 0.2 for the most challenging organ. Thus, we conclude that there is a lack of accurate and computationally light general-purpose segmentation method that can analyse dynamic total-body PET images.

Abstract Consumer wearables and sensors are a rich source of data about patients’ daily disease and symptom burden, particularly in the case of movement disorders like Parkinson’s disease (PD). However, interpreting these complex data into so-called digital biomarkers requires complicated analytical approaches, and validating these biomarkers requires sufficient data and unbiased evaluation methods. Here we describe the use of crowdsourcing to specifically evaluate and benchmark features derived from accelerometer and gyroscope data in two different datasets to predict the presence of PD and severity of three PD symptoms: tremor, dyskinesia, and bradykinesia. Forty teams from around the world submitted features, and achieved drastically improved predictive performance for PD status (best AUROC = 0.87), as well as tremor- (best AUPR = 0.75), dyskinesia- (best AUPR = 0.48) and bradykinesia-severity (best AUPR = 0.95).

Abstract Background We investigated the image quality of 11C, 68Ga, 18F and 89Zr, which have different positron fractions, physical half-lifes and positron ranges. Three small animal positron emission tomography/computed tomography (PET/CT) systems were used in the evaluation, including the Siemens Inveon, RAYCAN X5 and Molecubes β-cube. The evaluation was performed on a single scanner level using the national electrical manufacturers association (NEMA) image quality phantom and analysis protocol. Acquisitions were performed with the standard NEMA protocol for 18F and using a radionuclide-specific acquisition time for 11C, 68Ga and 89Zr. Images were assessed using percent recovery coefficient (%RC), percentage standard deviation (%STD), image uniformity (%SD), spill-over ratio (SOR) and evaluation of image quantification. Results 68Ga had the lowest %RC ( 85%) and lowest %STD for the 5 mm rod across all systems. For 11C and 89Zr, the maximum %RC was close (> 76%) to the %RC with 18F. A larger SOR were measured in water with 11C and 68Ga compared to 18F on all systems. SOR in air reflected image reconstruction and data correction performance. Large variation in image quantification was observed, with maximal errors of 22.73% (89Zr, Inveon), 17.54% (89Zr, RAYCAN) and − 14.87% (68Ga, Molecubes). Conclusions The systems performed most optimal in terms of NEMA image quality parameters when using 18F, where 11C and 89Zr performed slightly worse than 18F. The performance was least optimal when using 68Ga, due to large positron range. The large quantification differences prompt optimization not only by terms of image quality but also quantification. Further investigation should be performed to find an appropriate calibration and harmonization protocol and the evaluation should be conducted on a multi-scanner and multi-center level.

New SARS-CoV-2 variants, breakthrough infections, waning immunity, and sub-optimal vaccination rates account for surges of hospitalizations and deaths. There is an urgent need for clinically valuable and generalizable triage tools assisting the allocation of hospital resources, particularly in resource-limited countries. We developed and validate CODOP, a machine learning-based tool for predicting the clinical outcome of hospitalized COVID-19 patients. CODOP was trained, tested and validated with six cohorts encompassing 29223 COVID-19 patients from more than 150 hospitals in Spain, the USA and Latin America during 2020–22. CODOP uses 12 clinical parameters commonly measured at hospital admission for reaching high discriminative ability up to 9 days before clinical resolution (AUROC: 0·90–0·96), it is well calibrated, and it enables an effective dynamic risk stratification during hospitalization. Furthermore, CODOP maintains its predictive ability independently of the virus variant and the vaccination status. To reckon with the fluctuating pressure levels in hospitals during the pandemic, we offer two online CODOP calculators, suited for undertriage or overtriage scenarios, validated with a cohort of patients from 42 hospitals in three Latin American countries (78–100% sensitivity and 89–97% specificity). The performance of CODOP in heterogeneous and geographically disperse patient cohorts and the easiness of use strongly suggest its clinical utility, particularly in resource-limited countries.

Abstract Background Prognostic markers specific to a particular cancer type can assist in the evaluation of survival probability of patients and help clinicians to assess the available treatment modalities. Methods Gene expression data was analyzed from three independent colon cancer microarray gene expression data sets (N = 1052). Survival analysis was performed for the three data sets, stratified by the expression level of the LINE-1 type transposase domain containing 1 (L1TD1). Correlation analysis was performed to investigate the role of the interactome of L1TD1 in colon cancer patients. Results We found L1TD1 as a novel positive prognostic marker for colon cancer. Increased expression of L1TD1 associated with longer disease-free survival in all the three data sets. Our results were in contrast to a previous study on medulloblastoma, where high expression of L1TD1 was linked with poor prognosis. Notably, in medulloblastoma L1TD1 was co-expressed with its interaction partners, whereas our analysis revealed lack of co-expression of L1TD1 with its interaction partners in colon cancer. Conclusions Our results identify increased expression of L1TD1 as a prognostic marker predicting longer disease-free survival in colon cancer patients.

Background:. Because of the increasing number of total hip arthroplasties (THAs), even a small proportion of complications after the operation can lead to substantial individual difficulties and health-care costs. The aim of this study was to develop simple-to-use risk prediction models to assess the risk of the most common reasons for implant failure to facilitate clinical decision-making and to ensure long-term survival of primary THAs. Methods:. We analyzed patient and surgical data reported to the Finnish Arthroplasty Register (FAR) on 25,919 primary THAs performed in Finland between May 2014 and January 2018. For the most frequent adverse outcomes after primary THA, we developed multivariable Lasso regression models based on the data of the randomly selected training cohort (two-thirds of the data). The performances of all models were validated using the remaining, independent test set consisting of 8,640 primary THAs (one-third of the data) not used for building the models. Results:. The most common outcomes within 6 months after the primary THA were revision operations due to periprosthetic joint infection (1.1%), dislocation (0.7%), or periprosthetic fracture (0.5%), and death (0.7%). For each of these outcomes, Lasso regression identified subsets of variables required for accurate risk predictions. The highest discrimination performance, in terms of area under the receiver operating characteristic curve (AUROC), was observed for death (0.84), whereas the performance was lower for revisions due to periprosthetic joint infection (0.68), dislocation (0.64), or periprosthetic fracture (0.65). Conclusions:. Based on the small number of preoperative characteristics of the patient and modifiable surgical parameters, the developed risk prediction models can be easily used to assess the risk of revision or death. All developed models hold the potential to aid clinical decision-making, ultimately leading to improved clinical outcomes. Level of Evidence:. Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.

We studied whether TOF reduces error propagation from attenuation correction to PET image reconstruction in PET/MR neuroimaging, by using imperfect attenuation maps in a clinical PET/MR system with 525 ps timing resolution. Ten subjects who had undergone 18F-FDG PET neuroimaging were included. Attenuation maps using a single value (0.100 cm−1) with and without air, and a 3-class attenuation map with soft tissue (0.096 cm−1), air and bone (0.151 cm−1) were used. CT-based attenuation correction was used as a reference. Volume-of-interest (VOI) analysis was conducted. Mean bias and standard deviation across the brain was studied. Regional correlations and concordance were evaluated. Statistical testing was conducted. Average bias and standard deviation were slightly reduced in the majority (23–26 out of 35) of the VOI with TOF. Bias was reduced near the cortex, nasal sinuses, and in the mid-brain with TOF. Bland–Altman and regression analysis showed small improvements with TOF. However, the overall effect of TOF to quantitative accuracy was small (3% at maximum) and significant only for two attenuation maps out of three at 525 ps timing resolution. In conclusion, TOF might reduce the quantitative errors due to attenuation correction in PET/MR neuroimaging, but this effect needs to be further investigated on systems with better timing resolution.

The response to respiratory virus exposure can currently not be predicted by pre- or early post-exposure molecular signatures. Here, the authors conduct a community-based analysis of blood gene expression from healthy individuals exposed to respiratory viruses and provide predictive models and biological insight into the physiological response.

In this review, we will summarize the past and current state-of-the-art developments in attenuation and scatter correction approaches for hybrid positron emission tomography (PET) and magnetic resonance (MR) imaging. The current status of the methodological advances for producing accurate attenuation and scatter corrections on PET/MR systems are described, in addition to emerging clinical and research applications. Future prospects and potential applications that benefit from accurate data corrections to improve the quantitative accuracy and clinical applicability of PET/MR are also discussed. Novel clinical and research applications where improved attenuation and scatter correction methods are beneficial are highlighted.

We present a novel method for estimating respiratory motion using inertial measurement units (IMUs) based on microelectromechanical systems (MEMS) technology. As an application of the method we consider the amplitude gating of positron emission tomography (PET) imaging, and compare the method against a clinically used respiration motion estimation technique. The presented method can be used to detect respiratory cycles and estimate their lengths with state-of-the-art accuracy when compared to other IMU-based methods, and is the first based on commercial MEMS devices, which can estimate quantitatively both the magnitude and the phase of respiratory motion from the abdomen and chest regions. For the considered test group consisting of eight subjects with acute myocardial infarction, our method achieved the absolute breathing rate error per minute of 0.44 ± 0.23 1/min, and the absolute amplitude error of 0.24 ± 0.09 cm, when compared to the clinically used respiratory motion estimation technique. The presented method could be used to simplify the logistics related to respiratory motion estimation in PET imaging studies, and also to enable multi-position motion measurements for advanced organ motion estimation.

Magnetic resonance imaging-only radiotherapy treatment planning (MRI-only RTP) and positron emission tomography (PET)–MRI imaging require generation of synthetic computed tomography (sCT) images from MRI images. In this study, initial dosimetric evaluation was performed for a previously developed MRI-based attenuation correction (MRAC) method for use in MRI-only RTP of the brain. MRAC-based sCT images were retrospectively generated from Dixon MR images of 20 patients who had previously received external beam radiation therapy (EBRT). Bone segmentation performance and Dice similarity coefficient of the sCT conversion method were evaluated for bone volumes on CT images. Dose calculation accuracy was assessed by recalculating the CT-based EBRT plans using the sCT images as the base attenuation data. Dose comparison was done for the sCT- and CT-based EBRT plans in planning target volume (PTV) and organs at risk (OAR). Parametric dose comparison showed mean relative differences of

Pathway analysis is a common approach in diverse biomedical studies, yet the currently-available pathway tools do not typically support the increasingly popular personalized analyses. Another weakness of the currently-available pathway methods is their inability to handle challenging data with only modest group-based effects compared to natural individual variation. In an effort to address these issues, this study presents a novel pathway method PASI (Pathway Analysis for Sample-level Information) and demonstrates its performance on complex diseases with different levels of group-based differences in gene expression. PASI is freely available as an R package.

Dual cardiac and respiratory gating is a well-known technique for motion compensation in nuclear medicine imaging. In this study, we present a new data fusion framework for dual cardiac and respiratory gating based on multidimensional microelectromechanical (MEMS) motion sensors. Our approach aims at robust estimation of the chest vibrations, that is, high-frequency precordial vibrations and low-frequency respiratory movements for prospective gating in positron emission tomography (PET), computed tomography (CT), and radiotherapy. Our sensing modality in the context of this paper is a single dual sensor unit, including accelerometer and gyroscope sensors to measure chest movements in three different orientations. Since accelerometer- and gyroscope-derived respiration signals represent the inclination of the chest, they are similar in morphology and have the same units. Therefore, we use principal component analysis (PCA) to combine them into a single signal. In contrast to this, the accelerometer- and gyroscope-derived cardiac signals correspond to the translational and rotational motions of the chest, and have different waveform characteristics and units. To combine these signals, we use independent component analysis (ICA) in order to obtain the underlying cardiac motion. From this cardiac motion signal, we obtain the systolic and diastolic phases of cardiac cycles by using an adaptive multi-scale peak detector and a short-time autocorrelation function. Three groups of subjects, including healthy controls (n = 7), healthy volunteers (n = 12), and patients with a history of coronary artery disease (n = 19) were studied to establish a quantitative framework for assessing the performance of the presented work in prospective imaging applications. The results of this investigation showed a fairly strong positive correlation (average r = 0.73 to 0.87) between the MEMS-derived (including corresponding PCA fusion) respiration curves and the reference optical camera and respiration belt sensors. Additionally, the mean time offset of MEMS-driven triggers from camera-driven triggers was 0.23 to 0.3 ± 0.15 to 0.17 s. For each cardiac cycle, the feature of the MEMS signals indicating a systolic time interval was identified, and its relation to the total cardiac cycle length was also reported. The findings of this study suggest that the combination of chest angular velocity and accelerations using ICA and PCA can help to develop a robust dual cardiac and respiratory gating solution using only MEMS sensors. Therefore, the methods presented in this paper should help improve predictions of the cardiac and respiratory quiescent phases, particularly with the clinical patients. This study lays the groundwork for future research into clinical PET/CT imaging based on dual inertial sensors.