Image Data Research Articles

POSSIBLE ERRORS IN THE DIAGNOSIS OF INTRAOCULAR TUMORS

Rationale: The article discusses difficulties in the differential diagnosis of intraocular tumors (IOT), particularly in cases of small choroidal melanoma. Despite a substantial number of publications dedicated to the diagnosis of intraocular tumors, difficulties in interpreting imaging methods, such as optical coherence tomography (OCT), persist.Objective: to analyze the OCT patterns that have led to diagnostic errors.Methods: We analyzed the OCT results of 15 patients whose diagnosis of «choroidal hemangioma» was revised to «choroidal melanoma» in 11 cases or «progressive nevus» in 4 cases. The mean age at the time of IOT detection was 50.20±13.08 years (range: 35–79 years); the average tumor prominence was 2.22±0.82 mm (range: 1-3.6 mm), and the average diameter was 8.01±2.72 mm (range: 3.11–11.5 mm). A detailed analysis of the specific OCT patterns that led to diagnostic errors was conducted.Results and discussion: The characteristics of the identified OCT patterns that frequently caused diagnostic difficulties were thoroughly examined, such as cystic changes and the «subretinal cleft» phenomenon. Cases in which complex OCT patterns overlapped with both melanoma and hemangioma were also analyzed, reinforcing the need for a multimodal approach to the interpretation of imaging data. Conclusion: for accurate diagnosis and to avoid erroneous planning of treatment strategies, OCT should be evaluated within the context of the clinical picture and patient history.

Dual-encoder architecture for metal artifact reduction for kV-cone-beam CT images in head and neck cancer radiotherapy.

During a radiotherapy(RT) course, geometrical variations of target volumes, organs at risk, weight changes (loss/gain), tumor regression and/or progression can significantly affect the treatment outcome. Adaptive RT has become the effective methods along with technical advancements in imaging modalities including cone-beam computed tomography (CBCT). Planning CT (pCT) can be modified via deformable image registration (DIR), which is applied to the pair of pCT and CBCT. However, the artifact existed in both pCT and CBCT is a vulnerable factor in DIR. The dose calculation on CBCT is also suggested. Missing information due to the artifacts hinders the accurate dose calculation on CBCT. In this study, we aim to develop a deep learning-based metal artifact reduction(MAR) model to reduce the metal artifacts in CBCT for head and neck cancer RT. To train the proposed MAR model, we synthesized the kV-CBCT images including metallic implants, with and without metal artifacts (simulated image data pairs) through sinogram image handling process. We propose the deep learning architecture which focuses on both artifact removal and reconstruction of anatomic structure using a dual-encoder architecture. We designed four single-encoder models and three dual-encoder models based on UNet (for an artifact removal) and FusionNet (for a tissue restoration). Each single-encoder model contains either UNet or FusionNet, while the dual-encoder models have both UNet and FusionNet architectures. In the dual-encoder models, we implemented different feature fusion methods, including simple addition, spatial attention, and spatial/channel wise attention. Among the models, a dual-encoder model with spatial/channel wise attention showed the highest scores in terms of peak signal-to-noise ratio, mean squared error, structural similarity index, and Pearson correlation coefficient. CBCT images from 34 head and neck cancer patients were used to test the developed models. The dual-encoder model with spatial/channel wise attention showed the best results in terms of artifact index. By using the proposed model to CBCT, one can achieve more accurate synthetic pCT for head and neck patients as well as better tissue recognition and structure delineation for CBCT image itself.

Psoas muscle index in sarcopenia following transjugular intrahepatic portosystemic shunt: A multicenter, retrospective study

AbstractAimsAlthough the skeletal muscle index at the third lumbar vertebra (L3‐SMI) is commonly utilized for the diagnosis of sarcopenia, the psoas muscle index at L3 (L3‐PMI) may serve as a reliable alternative indicator. This study aims to investigate the application of the PMI in patients who have undergone transjugular intrahepatic portosystemic shunt (TIPS).MethodsThis study included a cohort of 406 patients with cirrhosis who underwent TIPS between February 2016 and July 2022 across three medical centers in China. Clinical and imaging data, specifically L3‐SMI and L3‐PMI, were collected for these patients. The prognosis of the patients was assessed through re‐examinations and telephone follow‐ups, which extended up to 5 years. The diagnostic thresholds for sarcopenia, as defined by L3‐PMI and L3‐SMI, were established at 6.36 or 42.00 cm²/m² for males and 3.92 or 38.00 cm²/m² for females, respectively. Cox proportional hazards and Kaplan–Meier (K‐M) analyses were employed to evaluate patient survival.ResultsThe findings indicated that both L3‐PMI and L3‐SMI before TIPS were independent risk factors for mortality. The results of the paired t‐test demonstrated a significant increase in L3‐PMI 1 month post‐TIPS (5.11 ± 1.81 vs. 5.71 ± 1.90 cm²/m², p < 0.001), whereas L3‐SMI did not exhibit a significant increase until 6 months post‐TIPS (45.45 ± 9.41 vs. 48.59 ± 10.38 cm²/m², p < 0.001). Among these patients, 191 (47.0%) and 159 (39.2%) were diagnosed with sarcopenia according to the L3‐PMI and L3‐SMI models, respectively. Patients identified as sarcopenic by both indicators demonstrated a significantly lower survival rate (L3‐SMI model: hazard ratio [HR], 1.913; 95% confidence interval [CI], 1.094–3.410; log‐rank p = 0.020; L3‐PMI model: HR, 1.867; 95% CI, 1.059–3.290; log‐rank p = 0.030). In sarcopenic patients, the reversal of sarcopenia associated with improved survival occurred 1 month after TIPS in the L3‐PMI model (HR, 2.675; 95% CI, 1.245–5.735; log‐rank p = 0.012), while a similar effect was not observed until 6 months post‐TIPS in the L3‐SMI model (HR, 3.342; 95% CI, 1.477–7.560; log‐rank p = 0.004).ConclusionsL3‐PMI and L3‐SMI demonstrate comparable efficacy in diagnosing cirrhosis‐related sarcopenia. Furthermore, L3‐PMI has the capacity to identify improvements in sarcopenia as early as 1 month following TIPS, thereby providing earlier insights into patient survival outcomes.

A systematic review on feature extraction methods and deep learning models for detection of cancerous lung nodules at an early stage -The Recent trends and challenges.

Lung cancer is one of the most common life-threatening worldwide cancers affecting both the male and female populations. The appearance of nodules in the scan image is an early indication of the development of cancer cells in the lung. The Low Dose Computed Tomography screening technique is used for the early detection of cancer nodules. Therefore, with more Computed Tomography (CT) lung profiles, an automated lung nodule analysis system can be utilized through image processing techniques and neural network algorithms. A CT image of the lung consists of many elements such as blood vessels, ribs, nodules, sternum, bronchi, and nodules. These nodules can be both benign and malignant, where the latter leads to lung cancer. Detecting them at an earlier stage can increase life expectancy by up to 5 to 10 years. To analyze only the nodules from the profile, respected features are calculated using image processing techniques. Based on the review, textural features were promising ones in medical image analysis and for solving computer vision problems. The significance of extracting the hidden features (textural) enables the higher performance of deep learning algorithms (DL) predominantly in medical imaging, which has resulted in improvement in accuracy in recent years. Earlier detection of cancerous lung nodules is possible through the combination of multi-featured extraction and classification techniques using image data. In this paper, we discussed the overview of lung cancer along with publicly available datasets for research purposes. The primary objective of the paper is to provide the importance of textural features when combined with different deep-learning models. It gives insights into their advantages, disadvantages, and limitations regarding possible research gaps. The paper compared the recent studies of deep learning models with and without feature extraction and concluded that DL models that include feature extraction were better than the others.

Enhancing image-based diagnosis of gastrointestinal tract diseases through deep learning with EfficientNet and advanced data augmentation techniques.

The early detection and diagnosis of gastrointestinal tract diseases, such as ulcerative colitis, polyps, and esophagitis, are crucial for timely treatment. Traditional imaging techniques often rely on manual interpretation, which is subject to variability and may lack precision. Current methodologies leverage conventional deep learning models that, while effective to an extent, often suffer from overfitting and generalization issues on medical image datasets due to the intricate and subtle variations in disease manifestations. These models typically do not fully utilize the potential of transfer learning or advanced data augmentation, leading to less-than-optimal performance, especially in diverse real-world scenarios where data variability is high. This study introduces a robust model using the EfficientNetB5 architecture combined with a sophisticated data augmentation strategy. The model is tailored for the high variability and intricate details present in gastrointestinal tract disease images. By integrating transfer learning with maximal pooling and extensive regularization, the model aims to enhance diagnostic accuracy and reduce overfitting. The proposed model achieved a test accuracy of 98.89%, surpassing traditional methods by incorporating advanced regularization and augmentation techniques. The application of horizontal flipping and dynamic scaling during training significantly improved the model's ability to generalize, evidenced by a low-test loss of 0.230 and high precision metrics across all classes. The proposed deep learning framework demonstrates superior performance in the automated classification of gastrointestinal diseases from image data. By addressing key limitations of existing models through innovative techniques, this study contributes to the enhancement of diagnostic processes in medical imaging, potentially leading to more accurate and timely disease interventions.

Novel Algorithm for Surgical Management of Cervical Ossification of Posterior Longitudinal Ligament: A Retrospective Cohort Study With 2-Year Follow-Up.

With the development of new technologies, the surgical algorithm for ossification of posterior longitudinal ligament (OPLL) in the cervical spine also needs to be updated. The aim of this study is to elucidate a new surgical classification algorithm to guide the choice of surgical approaches for cervical OPLL based on its location and extent. In this algorithm, anterior controllable antedisplacement and fusion (ACAF) will be used as a new surgical option. This is a single-centered, retrospective, cohort study utilizing a novel algorithm based on the following three criteria: (1) the positional relationship between ossification and uncinate process (UP), (2) the K-line, and (3) the ossification segment for surgical decision-making. Patients diagnosed with cervical OPLL who received surgical intervention guided by the algorithm were included. Patient demographics, Japanese Orthopedic Association (JOA) scores, surgical time, imaging data before and 2 years after surgery, and the occurrence of complications were extracted from the database. Paired t-test was used for intragroup comparison, and one-way ANOVA test was used for the intergroup analyses. Based on this novel algorithm, 15 patients with Type I, 8 patients with Type II a, 2 patients with Type II b1, 5 patients with Type II b2 were included. The decision-making for the surgical techniques used in each patient followed the recommendation of the novel algorithm. The postoperative JOA scores of all types of patients improved significantly (p < 0.05), and the invasion rates of vertebral canal had also been significantly reduced (p < 0.05). In terms of restoring cervical curvature, patients with Type I (receiving ACAF) and Type II b2 (receiving laminectomy with instrumented fusion) benefited more from surgery (p < 0.05). A new algorithm guiding the choice of surgical approach for cervical OPLL was validated in a series of 30 patients. Through this analysis, we obtained on their clinical outcomes and complications. ACAF surgery is an ideal choice for Type I patients, with ossification located between UPs, while for patients with ossification exceeding UPs, it is better to perform anterior cervical corpectomy and fusion or posterior surgery.

Abstract 4119052: Redefining the phenotypic continuum of cardiovascular disease using machine learning

Background: Current approaches that define cardiovascular disease rely on assignment of traditional diagnostic labels which may not reflect the natural continuum of phenotypic expression, influenced by genetic and environmental modifiers, or adequately identify groups with shared molecular mechanisms. Machine learning can leverage advances in genetic and imaging characterisation to reclassify phenotypic diversity along a continuum, from health to disease. Research Aims: Our study aims to build a tree-like classification of cardiovascular phenotypes in the community where branches represent subjects with shared features, ordered by their severity. Using dilated cardiomyopathy (DCM) and hypertrophic cardiomyopathy (HCM) as exemplar conditions with opposing phenotypes, we aim to demonstrate the interaction between genetic and environmental modifiers of disease expression. Methods: We analysed cardiovascular magnetic resonance (CMR) imaging, electrocardiogram (ECG), biomarkers and clinical data from participants in the UK Biobank. Using unsupervised learning of multiparametric data, we built a tree of phenotypic expression. We projected the risk of DCM and HCM cases, and associated rare and common variant risk, onto the tree structure. Results: Ten main branches were discovered, using data from 41,525 participants (51.7 % female, median age 65 yrs [IQR: 58, 70]). The extremities of branches 1 and 6 were enriched for DCM cases (p &lt;0.05) and participants with reduced ejection fraction and high left ventricular volumes (p &lt;0.05). Compared with subjects at the tree centre, distal participants in branch 1 had increased blood pressure (BP) and body mass index (BMI), whilst those in branch 6 were more likely to have high HbA1c levels and carry a pathogenic DCM variant (p &lt;0.05). The extremity of branch 5 was enriched for HCM cases, rare sarcomeric variants (p = 0.004) and high polygenic risk (p &lt;0.001). Whilst the ends of branches 3 and 9 had high polygenic risk for HCM, they were less likely to have elevated BP and BMI, and had reduced likelihood of HCM expression, compared with branch 5. A phenomapping model was trained to accurately project unseen subjects onto the tree (R 2 0.98). Conclusions: We present a tree-like continuum of cardiovascular phenotypes, providing a novel framework for mechanistic discovery and exploration of genotype-phenotype associations. Phenomapping of unseen subjects enables personalised estimation of genetic and cardiovascular risk.

CPAP delivered via a helmet interface in lightly sedated patients with moderate to severe ARDS: predictors of success outside the ICU

Objective: This study aimed to describe the outcomes and explore predictors of intubation and mortality in patients with ARDS due to COVID-19 treated with CPAP delivered via a helmet interface and light sedation. Methods: This was a retrospective cohort study involving patients with COVID-19-related ARDS who received CPAP using a helmet developed in Brazil (ELMOT), associated with a light sedation protocol in a pulmonology ward. Demographic, clinical, imaging, and laboratory data, as well as the duration and response to the ELMO-CPAP sessions, were analyzed. Results: The sample comprised 180 patients. The intubation avoidance rate was 72.8%. The lack of necessity for intubation was positively correlated with younger age, &gt; 24-h continuous HELMET-CPAP use in the first session, &lt; 75% pulmonary involvement on CT, and ROX index &gt; 4.88 in the second hour. The overall in-hospital mortality rate was 18.9%, whereas those in the nonintubated and intubated groups were 3.0% and 61.2%, respectively. Advanced age increased the mortality risk by 2.8 times, escalating to 13 times post-intubation. Conclusions: ELMO-CPAP with light sedation in a pulmonology ward was successful in &gt; 70% of patients with moderate to severe ARDS due to COVID-19. Younger age, pulmonary involvement, ROX index, and prolonged first Helmet-CPAP session duration were associated with no need for intubation. Older age and intubation are associated with mortality.

Energy Associated With Dynamic Network Changes in Patients With Multiple Sclerosis and Cognitive Impairment.

Patients with multiple sclerosis (MS) often experience cognitive impairment, and this is related to structural disconnection and subsequent functional reorganization. It is unclear how specific patterns of functional reorganization might make it harder for cognitively impaired (CI) patients with MS to dynamically adapt how brain regions communicate, which is crucial for normal cognition. We aimed to identify dynamic functional network patterns that are relevant to cognitive impairment in MS and investigate whether these patterns can be explained by altered energy costs. Resting-state functional and diffusion MRI was acquired in a cross-sectional design, as part of the Amsterdam MS cohort. Patients with clinically definitive MS (relapse-free) were classified as CI (≥2/7 domains Z < -2), mildly CI (MCI) (≥2/7 domains Z < -1.5), or cognitively preserved (CP) based on an expanded Brief Repeatable Battery of Neuropsychological Tests. Functional connectivity states were determined using k-means clustering of moment-to-moment cofluctuations (i.e., edge time series), and the resulting state sequence was used to characterize the frequency of transitions. Control energy of the state transitions was calculated using the structural network with network control theory. Imaging and cognitive data were available for 95 controls and 330 patients (disease duration: 15 years; 179 CP, 65 MCI, and 86 CI). We identified a "visual network state," "sensorimotor network state," "ventral attention network state," and "default mode network state." CI patients transitioned less frequently between connectivity states compared with CP (β = -5.78; p = 0.038). Relative to the time spent in a state, CI patients transitioned less from a "default mode network state" to a "visual network state" (β = -0.02; p = 0.004). The CI patients required more control energy to transition between states (β = 0.32; p = 0.007), particularly for the same transition (β = 0.34; p = 0.049). This study showed that it costs more energy for MS patients with cognitive impairment to dynamically change the functional network, possibly explaining why these transitions occur less frequently. In particular, transitions from a default mode network state to a visual network state were relevant for cognition in these patients. To further study the order of events leading to these network disturbances, future work should include longitudinal data across different disease stages.

Abstract 4121793: A Pressure-Volume Loops Approach Predicts Outcomes After Double Switch Operation For Congenitally Corrected Transposition Of The Great Arteries with Intact Ventricular Septum

Background: Double switch operation (DSO) for congenitally corrected transposition of the great arteries with intact ventricular septum (ccTGA) has a high rate of post-DSO LV dysfunction. Hypothesis: LV pressure-volume area (PVA), a surrogate of myocardial O 2 consumption, is a superior marker of LV preparedness and is associated with adverse outcomes after DSO. Aims: 1. Derive a mathematical relationship to estimate LV PVA (ePVA) from catheterization (pressure) and imaging (volumetric) data, using directly measured PVA (mPVA) as reference 2. Assess if lower ratio of LV ePVA to RV ePVA (as an internal control) is associated with adverse outcome after DSO. Methods: Aim 1: Using conductance catheter derived invasive subpulmonary LV PV loops, mPVA was recorded as the sum of stroke work (SW) and potential energy (PE). A mathematical relationship was established with standard catheterization/imaging data to estimate ePVA. Aim 2: In a retrospective cohort, LV:RV ePVA ratio was calculated as above, and along with standard clinical metrics, assessed for relationship with a composite outcome of ≥ moderate LV dysfunction, transplant, or death post-DSO. Results: Aim 1: In 20 PV loop studies in 18 pre-DSO patients, there was a strong linear correlation between measured and estimated SW and PE (R 2 &gt; 0.9 and p&lt;0.0001 for both). ePVA yielded high agreement and low bias compared to mPVA (mean bias 0.5±11%). Aim 2: Composite outcome occurred in 6/42 DSO patients (14%). LV:RV ePVA ratio (0.57 [0.49, 0.61] vs 0.90 [0.73, 1.1], p&lt;0.001) and LV:RV pressure ratio (0.80 [0.77, 0.87] vs 1.13 [1.01, 1.23], p&lt;0.001) were lower in those with adverse outcome. There were no differences in other pre-operative parameters. On time to event analysis, lower LV:RV ePVA ratio was the strongest determinant (HR 39, 95% CI 15-100; C Index 0.94), while lower LV:RV pressure ratio was the only other predictor (HR 24, 95% CI 13-42; C Index 0.88). In 8 patients with borderline pressure ratios of 0.77-0.88, ePVA ratio was an excellent discriminator - 3 patients with ePVA ratio of ≥0.67 had good outcomes, whereas 5 patients with ePVA ratio &lt;0.67 had adverse outcomes. Conclusion: LV:RV ePVA ratio &gt;0.67 is a strong and novel predictor of LV preparedness for DSO in patients with ccTGA.

Enhancing Lung and Breast Cancer Screening with Advanced AI and Image Processing Techniques

This research investigates the application of CNNs for diagnostics improvements in lung and breast cancers based on AI image classification approaches. Using the datasets with 15,000 images describing lung cancer and 10,000 images describing cases of breast cancer, the models showed high performance: 90% for lung cancer and 99% for breast cancer classification. The descriptive analysis pointed out different features in imaging, such as dense tissue structure and irregular cell patterns; the models successfully identified these. The findings underlined the vital role that AI could play in assisting radiologists by delivering preliminary analysis, triaging high-risk cases, and leading to early cancer detection. Essential challenges were highlighted: ethical considerations concerning patients' privacy and AI algorithms' transparency. The limitation of the dataset diversity resulted in the conclusion that only broader data can ensure good generalization in various clinical settings. They recommended integrating the AI tool with clinical workflow and also called for training radiologists for effectiveness. Future research directions include real-time imaging and patient data integration for comprehensive diagnostic support and multi-modal approaches that combine imaging with genomics for more precise predictions. This leads to a more personalized cancer diagnosis and treatment plan, thus ultimately improving the results of the patients. This research, therefore, underlines the transformative capabilities of AI and image processing in modernizing cancer screening and diagnostics toward more accurate and efficient healthcare practices.

Dyspnea in COVID-associated lung disease

Introduction. Covid-associated lung disease has become one of the leading problems of the COVID-19 pandemic, and early diagnosis of complications is complicated. Assessment of dyspnoea as a significant symptom is important, but its diagnostic ability in this lesion is poorly understood.Aim. To study the features of dyspnoea in COVID-19-associated lung lesions and its diagnostic value.Materials and methods. The study included 134 patients with COVID-19-confirmed pneumonia. Demographic and anthropometric data, subjective condition, dyspnoea severity according to Borg scale, concomitant pathologies, multispiral computed tomography (MSCT) data, arterial blood gas composition, capnometry and spirometry data, as well as disease outcomes (transfer to ICU, support ventilation, fatal outcome) were analysed.Results. Dyspnoea was present in 43.3% of patients. Increased dyspnoea was associated with increased C-reactive protein (CRP), D-dimer, lung tissue damage (MSCT), decreased forced vital capacity (FVC), and increased alveolar-arterial gradient (P(A-a)O2). Dyspnoea correlated with duration of hospitalisation and need for oxygen therapy (OR = 1.307, p = 0.008). 57.4% of patients with hypoxaemia did not complain of dyspnoea, but their outcomes did not differ between patients with dyspnoea and hypoxaemia. 32.2% of patients without hypoxaemia complained of dyspnoea. These patients did not have a significant increase in FGEF by the time of discharge, and it remained lower than in patients without dyspnoea and hypoxaemia.Conclusion. Dyspnoea in COVID-19-associated lung disease is an important symptom correlating with clinical-functional, instrumental and laboratory characteristics of the disease. Comprehensive data analysis is necessary to identify patients requiring further observation.

Comparison and identification of human coronary plaques with/without erosion using patient-specific optical coherence tomography-based fluid-structure interaction models: a pilot study.

Plaque erosion (PE) with secondary thrombosis is one of the key mechanisms of acute coronary syndrome (ACS) which often leads to drastic cardiovascular events. Identification and prediction of PE are of fundamental significance for disease diagnosis, prevention and treatment. In vivo optical coherence tomography (OCT) data of eight eroded plaques and eight non-eroded plaques were acquired to construct three-dimensional fluid-structure interaction models and obtain plaque biomechanical conditions for investigation. Plaque stenosis severity, plaque burden, plaque wall stress (PWS) and strain (PWSn), flow shear stress (FSS), and ΔFSS (FSS variation in time) were extracted for comparison and prediction. A logistic regression model was used to predict plaque erosion. Our results indicated that the combination of mean PWS and mean ΔFSS gave best prediction (AUC = 0.866, 90% confidence interval (0.717, 1.0)). The best single predictor was max ΔFSS (AUC = 0.819, 90% confidence interval (0.624, 1.0)). The average of maximum FSS values from eroded plaques was 76% higher than that from the non-eroded plaques (127.96 vs. 72.69 dyn/cm2) while the average of mean FSS from erosion sites of the eight eroded plaques was 48.6% higher than that from sites without erosion (71.52 vs. 48.11 dyn/cm2). The average of mean PWS from plaques with erosion was 22.83% lower than that for plaques without erosion (83.2kPa vs. 107.8kPa). This pilot study suggested that combining plaque stress, strain and flow shear stress could help better identify patients with potential plaque erosion, enabling possible early intervention therapy. Further studies are needed to validate our findings.

A CT-based radiomics analyses for differentiating drug‑resistant and drug-sensitive pulmonary tuberculosis.

To explore the value of computed tomography based radiomics in the differential diagnosis of drug-sensitive and drug-resistant pulmonary tuberculosis. The clinical and computed tomography image data of 177 patients who were diagnosed with pulmonary tuberculosis through sputum culture and completed drug-susceptibility testing from April 2018 to December 2020 at the Second Hospital of Nanjing were retrospectively analyzed. Patients with drug-resistant pulmonary tuberculosis (n = 78) and drug-sensitive pulmonary tuberculosis (n = 99) were randomly divided into a training set (n = 124) and a validation set (n = 53) at a ratio of 7:3. Regions of interest were drawn to delineate the lesions and radiomics features were extracted from non-contrast computed tomography images. A radiomics signature based on the valuable radiomics features was constructed and a radiomics score was calculated. Demographic data, clinical symptoms, laboratory results and computed tomography imaging characteristics were evaluated to establish a clinical model. Combined with the Rad-score and clinical factors, a radiomics-clinical model nomogram was constructed. Thirteen features were used to construct the radiomics signature. The radiomics signature showed good discrimination in the training set (area under the curve (AUC), 0.891; 95% confidence interval (CI), 0.832-0.951) and the validation set (AUC, 0.803; 95% CI, 0.674-0.932). In the clinical model, the AUC of the training set was 0.780(95% CI, 0.700-0.859), while the AUC of the validation set was 0.692 (95% CI, 0.546-0.839). The radiomics-clinical model showed good calibration and discrimination in the training set (AUC, 0.932;95% CI, 0.888-0.977) and the validation set (AUC, 0.841; 95% CI, 0.719-0.962). Simple radiomics signature is of great value in differentiating drug-sensitive and drug-resistant pulmonary tuberculosis patients. The radiomics-clinical model nomogram showed good predictive, which may help clinicians formulate precise treatments.

Spatially integrated cortico-subcortical tracing data for analyses of rodent brain topographical organization.

The cerebral cortex extends axonal projections to several subcortical brain regions, including the striatum, thalamus, superior colliculus, and pontine nuclei. Experimental tract-tracing studies have shown that these subcortical projections are topographically organized, reflecting the spatial organization of sensory surfaces and body parts. Several public collections of mouse- and rat- brain tract-tracing data are available, with the Allen mouse brain connectivity atlas being most prominent. There, a large body of image data can be inspected, but it is difficult to combine data from different experiments and compare spatial distribution patterns. To enable co-visualization and comparison of topographical organization in mouse brain cortico-subcortical projections across experiments, we represent axonal labelling data as point data in a common 3D brain atlas space. We here present a collection of point-cloud data representing spatial distribution of corticostriatal, corticothalamic, corticotectal, and corticopontine projections in mice and exemplify how these spatially integrated point data can be used as references for experimental investigations of topographic organization in transgenic mice, and for cross-species comparison with corticopontine projections in rats.

Changes in brain connectivity and neurovascular dynamics during dexmedetomidine-induced loss of consciousness.

Understanding the neurophysiological changes that occur during loss and recovery of consciousness is a fundamental aim in neuroscience and has marked clinical relevance. Here, we utilize multimodal magnetic resonance neuroimaging to investigate changes in regional network connectivity and neurovascular dynamics as the brain transitions from wakefulness to dexmedetomidine-induced unconsciousness, and finally into early-stage recovery of consciousness. We observed widespread decreases in functional connectivity strength across the whole brain, and targeted increases in structure-function coupling (SFC) across select networks -- especially the cerebellum -- as individuals transitioned from wakefulness to hypnosis. We also observed robust decreases in cerebral blood flow (CBF) across the whole brain -- especially within the brainstem, thalamus, and cerebellum. Moreover, hypnosis was characterized by significant increases in the amplitude of low-frequency fluctuations (ALFF) of the resting-state blood oxygen level-dependent signal, localized within visual and somatomotor regions. Critically, when transitioning from hypnosis to the early stages of recovery, functional connectivity strength and SFC -- but not CBF -- started reverting towards their awake levels, even before behavioral arousal. By further testing for a relationship between connectivity and neurovascular alterations, we observed that during wakefulness, brain regions with higher ALFF displayed lower functional connectivity with the rest of the brain. During hypnosis, brain regions with higher ALFF displayed weaker coupling between structural and functional connectivity. Correspondingly, brain regions with stronger functional connectivity strength during wakefulness showed greater reductions in CBF with the onset of hypnosis. Earlier recovery of consciousness was associated with higher baseline (awake) levels of functional connectivity strength, CBF, and ALFF, as well as female sex. Across our findings, we also highlight the role of the cerebellum as a recurrent marker of connectivity and neurovascular changes between states of consciousness. Collectively, these results demonstrate that induction of, and emergence from dexmedetomidine-induced unconsciousness are characterized by widespread changes in connectivity and neurovascular dynamics.

Reducing the Brace Correction Stress on the Secondary Lumbar Curve Results in Excellent Muscle, Bone, and Disc Mechanical Performance: A Musculoskeletal Finite Element Simulation of AIS Patient With Rigo A3.

The biomechanical mechanism of brace intervention on bone, muscle, and disc should be comprehensively considered for AIS patients. We aimed to developmentally construct a musculoskeletal finite element model of adolescent idiopathic scoliosis to simulate the coupling of corrective forces and analyze the mechanical properties of bone, muscle, and disc. Investigateing, more effective clinical interventions to break the vicious cycle of patients during growth. A finite element model, including muscle, bone, and disc, was established using computed tomography data of a patient with RigoA3 adolescent idiopathic scoliosis. The three-point force coupling, antigravity, and bending effects of the Chêneau brace were simulated, and the correction force of the secondary lumbar bend was gradually reduced while observing the mechanical characteristics of bone, muscle, and disc. The correction force in line with symmetrical spine growth was comprehensively evaluated. The correction rate of the main thoracic (MT) curve, the intervertebral space height on the concave side of the vertebrae at the apex, and the stress ratio of the intervertebral discs were optimal when the maximum corrective pressure threshold was reached. However, the proximal thoracic (PT) curve was aggravated and the axial forces on the concave side were unbalanced. At this time, the biomechanical performance of the model is also not optimal. The correction rate of the Cobb Angle of the MT curve decreased with the decrease of the correction pressure in the lumbar region. When reduced to 25% of the maximum threshold, the convex side of disc stress, intervertebral space, and muscle axial force is more in line with the biomechanical mechanism of correction and can avoid sacrificing the PT curve. Downward adjustment of the corrective force to the secondary lumbar curve, using the Chêneau brace, results in better primary thoracic curvature mechanics in the musculoskeletal finite element model, suggesting that breaking the vicious cycle of scoliosis progression to guide benign spinal growth is beneficial.

Advancing hyperspectral imaging techniques for root systems: a new pipeline for macro- and microscale image acquisition and classification.

Understanding the environmental impacts on root growth and root health is essential for effective agricultural and environmental management. Hyperspectral imaging (HSI) technology provides a non-destructive method for detailed analysis and monitoring of plant tissues and organ development, but unfortunately examples for its application to root systems and the root-soil interface are very scarce. There is also a notable lack of standardized guidelines for image acquisition and data analysis pipelines. This study investigated HSI techniques for analyzing rhizobox-grown root systems across various imaging configurations, from the macro- to micro-scale, using the imec VNIR SNAPSCAN camera. Focusing on three graminoid species with different root architectures allowed us to evaluate the influence of key image acquisition parameters and data processing techniques on the differentiation of root, soil, and root-soil interface/rhizosheath spectral signatures. We compared two image classification methods, Spectral Angle Mapper (SAM) and K-Means clustering, and two machine learning approaches, Random Forest (RF) and Support Vector Machine (SVM), to assess their efficiency in automating root system image classification. Our study demonstrated that training a RF model using SAM classifications, coupled with wavelength reduction using the second derivative spectra with Savitzky-Golay (SG) smoothing, provided reliable classification between root, soil, and the root-soil interface, achieving 88-91% accuracy across all configurations and scales. Although the root-soil interface was not clearly resolved, it helped to improve the distinction between root and soil classes. This approach effectively highlighted spectral differences resulting from the different configurations, image acquisition settings, and among the three species. Utilizing this classification method can facilitate the monitoring of root biomass and future work investigating root adaptations to harsh environmental conditions. Our study addressed the key challenges in HSI acquisition and data processing for root system analysis and lays the groundwork for further exploration of VNIR HSI application across various scales of root system studies. This work provides a full data analysis pipeline that can be utilized as an online Python-based tool for the semi-automated analysis of root-soil HSI data.

Risk factors for healing failure after arthroscopic rotator cuff repair in small to medium-sized tears: a retrospective cohort study

ObjectivesTo identify risk factors for tendon healing failure following arthroscopic rotator cuff repair (ARCR) in patients with small to medium-sized rotator cuff tears (RCTs).MethodsA retrospective study was conducted on 320 patients with RCTs who underwent arthroscopic repair between June 2018 and June 2021. All patients had at least 2 years of postoperative follow-up, with MRI scans at the final assessment. Based on MRI results, patients were categorized into the healing success group (Group A: types I–III) or the healing failure group (Group B: types IV–V). Variables associated with rotator cuff healing, including patient characteristics, baseline symptoms, imaging data, and surgery-related factors, were analyzed using univariate and multivariate logistic regression.ResultsHealing failure occurred in 54 of the 320 patients (16.9%). Functional status improved significantly across all patients (P &amp;lt; 0.05), irrespective of healing outcomes. Multifactorial analysis identified smoking (OR = 1.931, P = 0.028), diabetes (OR = 3.517, P = 0.038), lower bone mineral density (BMD) (OR = 1.551, P = 0.018), higher fatty infiltration (FI) (OR = 4.025, P = 0.009), and smaller acromiohumeral distance (AHD) (OR = 2.546, P = 0.006) as independent risk factors for healing failure.ConclusionsSmoking, diabetes, lower BMD, higher FI, and smaller AHD are independent risk factors for healing failure following ARCR.

SURG-19. THE SYNDROME OF THE SUPPLEMENTARY MOTOR AREA: CLINICAL COURSE AND PREDICTORS OF OUTCOME AFTER RESECTION OF PREMOTOR LESIONS

Abstract OBJECTIVE Supplementary motor area (SMA) syndrome frequently occurs after resection of premotor lesions. Interhemispheric connectivity including the contralateral SMA might be crucially involved in its development and recovery. We aimed to identify predictors of outcome by quantifying preoperative structural alterations in the non-affected contralateral hemisphere. METHODS We retrospectively searched our database for patients developing SMA syndrome after resection of premotor lesions between 2013 and 2023. Clinical characteristics and imaging data were collected. Structural network analysis was performed using the previously validated morphometric inverse divergence network. RESULTS 27 patients with a mean age of 39.2 ± 17.7 years developed SMA syndrome, including 13/27 patients (48.1%) with primary brain tumors and 14/27 patients (51.9%) with intractable lesional epilepsy. Postoperative SMA syndrome was commonly characterized by contralateral upper extremity hemiparesis (19/27 patients, 70.4%) and aphasia in case of dominant hemisphere involvement (13/13 patients, 100%). Tumor patients were less likely to fully recover from SMA syndrome after 7 days compared to epilepsy patients (2/13 [15.4%] versus 9/14 [64.3%] patients, p=0.018). Most patients (23/27, 85.2%) experienced full recovery of SMA syndrome after a median follow-up of 5 months (IR 3-16 months), with no difference between patients with tumors or epilepsy (p=0.999). Preoperative contralateral intra-hemispheric connectivity was similar between patients receiving tumor or focal resection (0.183 vs 0.177, p=0.295). However, imaging analyses demonstrated baseline affection of SMA connectivity in patients with brain tumors as characterized by reduced cortical thickness of the contralateral frontal lobe and the cingulate gyrus compared to epilepsy patients. CONCLUSION While full recovery from SMA syndrome is common, SMA syndrome is frequently less severe and more transient in patients with epilepsy. This might be due to affection of the contralateral SMA network in tumor patients due to the infiltrative nature of the disease. Imaging analyses might assist in identifying network affections at baseline.