Many Models, Little Adoption-What Accounts for Low Uptake of Machine Learning Models for Atrial Fibrillation Prediction and Detection?

(1) Background: Atrial fibrillation (AF) is a major risk factor for stroke and is often underdiagnosed, despite being present in 13-26% of ischemic stroke patients. Recently, a significant number of machine learning (ML)-based models have been proposed for AF prediction and detection for primary and secondary stroke prevention. However, clinical translation of these technological innovations to close the AF care gap has been scant. Herein, we sought to systematically examine studies, employing ML models to predict incident AF in a population without prior AF or to detect paroxysmal AF in stroke cohorts to identify key reasons for the lack of translation into the clinical workflow. We conclude with a set of recommendations to improve the clinical translatability of ML-based models for AF. (2) Methods: MEDLINE, Embase, Web of Science, Clinicaltrials.gov, and ICTRP databases were searched for relevant articles from the inception of the databases up to September 2022 to identify peer-reviewed articles in English that used ML methods to predict incident AF or detect AF after stroke and reported adequate performance metrics. The search yielded 2815 articles, of which 16 studies using ML models to predict incident AF and three studies focusing on ML models to detect AF post-stroke were included. (3) Conclusions: This study highlights that (1) many models utilized only a limited subset of variables available from patients' health records; (2) only 37% of models were externally validated, and stratified analysis was often lacking; (3) 0% of models and 53% of datasets were explicitly made available, limiting reproducibility and transparency; and (4) data pre-processing did not include bias mitigation and sufficient details, leading to potential selection bias. Low generalizability, high false alarm rate, and lack of interpretability were identified as additional factors to be addressed before ML models can be widely deployed in the clinical care setting. Given these limitations, our recommendations to improve the uptake of ML models for better AF outcomes include improving generalizability, reducing potential systemic biases, and investing in external validation studies whilst developing a transparent modeling pipeline to ensure reproducibility.

Differential privacy preserved federated learning for prognostic modeling in COVID-19 patients using large multi-institutional chest CT dataset.

BACKGROUND: Notwithstanding the encouraging results of previous studies reporting on the efficiency of deep learning (DL) in COVID-19 prognostication, clinical adoption of the developed methodology still needs to be improved. To overcome this limitation, we set out to predict the prognosis of a large multi-institutional cohort of patients with COVID-19 using a DL-based model. PURPOSE: This study aimed to evaluate the performance of deep privacy-preserving federated learning (DPFL) in predicting COVID-19 outcomes using chest CT images. METHODS: After applying inclusion and exclusion criteria, 3055 patients from 19 centers, including 1599 alive and 1456 deceased, were enrolled in this study. Data from all centers were split (randomly with stratification respective to each center and class) into a training/validation set (70%/10%) and a hold-out test set (20%). For the DL model, feature extraction was performed on 2D slices, and averaging was performed at the final layer to construct a 3D model for each scan. The DensNet model was used for feature extraction. The model was developed using centralized and FL approaches. For FL, we employed DPFL approaches. Membership inference attack was also evaluated in the FL strategy. For model evaluation, different metrics were reported in the hold-out test sets. In addition, models trained in two scenarios, centralized and FL, were compared using the DeLong test for statistical differences. RESULTS: The centralized model achieved an accuracy of 0.76, while the DPFL model had an accuracy of 0.75. Both the centralized and DPFL models achieved a specificity of 0.77. The centralized model achieved a sensitivity of 0.74, while the DPFL model had a sensitivity of 0.73. A mean AUC of 0.82 and 0.81 with 95% confidence intervals of (95% CI: 0.79-0.85) and (95% CI: 0.77-0.84) were achieved by the centralized model and the DPFL model, respectively. The DeLong test did not prove statistically significant differences between the two models (p-value = 0.98). The AUC values for the inference attacks fluctuate between 0.49 and 0.51, with an average of 0.50 ± 0.003 and 95% CI for the mean AUC of 0.500 to 0.501. CONCLUSION: The performance of the proposed model was comparable to centralized models while operating on large and heterogeneous multi-institutional datasets. In addition, the model was resistant to inference attacks, ensuring the privacy of shared data during the training process.

Prevalence and Determinants of Long-Term Post-COVID Conditions in the United States: 2022 Behavioral Risk Factor Surveillance System.

BACKGROUND: A significant proportion of COVID survivors experience lingering and debilitating symptoms following acute COVID-19 infection. According to the national research plan on long COVID, it is a national priority to identify the prevalence of post-COVID conditions and their associated factors. METHOD: We performed a cross-sectional analysis of the Prevention Behavioral Risk Factor Surveillance System (BRFSS) 2022, the largest continuously gathered health survey dataset worldwide by the Centers for Disease Control. After identifying individuals with a positive history of COVID-19, we grouped COVID-19 survivors based on whether they experienced long-term post-COVID conditions. Using survey-specific R packages, we compared the two groups' socio-demographics, comorbidities, and lifestyle-related factors. A logistic regression model was used to identify factors associated with post-COVID conditions. RESULTS: The overall estimated prevalence of long-term post-COVID conditions among COVID survivors was 21.7%. Fatigue (5.7%), dyspnea (4.2%), and anosmia/ageusia (3.8%) were the most frequent symptoms. Based on multivariate logistic regression analysis, female sex, body mass index (BMI)≥25, lack of insurance, history of pulmonary disease, depression, and arthritis, being a former smoker, and sleep duration <7 h/d were associated with higher odds of post-COVID conditions. On the other hand, age >64 y/o, Black race, and annual household income ≥$100k were associated with lower odds of post-COVID conditions. CONCLUSION: Our findings indicate a notable prevalence of post-COVID conditions, particularly among middle-aged women and individuals with comorbidities or adverse lifestyles. This high-risk demographic may require long-term follow-up and support. Further investigations are essential to facilitate the development of specified healthcare and therapeutic strategies for those suffering from post-COVID conditions.

Operational greenhouse-gas emissions of deep learning in digital pathology: a modelling study.

BACKGROUND: Deep learning is a promising way to improve health care. Image-processing medical disciplines, such as pathology, are expected to be transformed by deep learning. The first clinically applicable deep-learning diagnostic support tools are already available in cancer pathology, and their number is increasing. However, data on the environmental sustainability of these tools are scarce. We aimed to conduct an environmental-sustainability analysis of a theoretical implementation of deep learning in patient-care pathology. METHODS: For this modelling study, we first assembled and calculated relevant data and parameters of a digital-pathology workflow. Data were breast and prostate specimens from the university clinic at the Institute of Pathology of the Rheinisch-Westfälische Technische Hochschule Aachen (Aachen, Germany), for which commercially available deep learning was already available. Only specimens collected between Jan 1 and Dec 31, 2019 were used, to omit potential biases due to the COVID-19 pandemic. Our final selection was based on 2 representative weeks outside holidays, covering different types of specimens. To calculate carbon dioxide (CO2) or CO2 equivalent (CO2 eq) emissions of deep learning in pathology, we gathered relevant data for exact numbers and sizes of whole-slide images (WSIs), which were generated by scanning histopathology samples of prostate and breast specimens. We also evaluated different data input scenarios (including all slide tiles, only tiles containing tissue, or only tiles containing regions of interest). To convert estimated energy consumption from kWh to CO2 eq, we used the internet protocol address of the computational server and the Electricity Maps database to obtain information on the sources of the local electricity grid (ie, renewable vs non-renewable), and estimated the number of trees and proportion of the local and world's forests needed to sequester the CO2 eq emissions. We calculated the computational requirements and CO2 eq emissions of 30 deep-learning models that varied in task and size. The first scenario represented the use of one commercially available deep-learning model for one task in one case (1-task), the second scenario considered two deep-learning models for two tasks per case (2-task), the third scenario represented a future, potentially automated workflow that could handle 7 tasks per case (7-task), and the fourth scenario represented the use of a single potential, large, computer-vision model that could conduct multiple tasks (multitask). We also compared the performance (ie, accuracy) and CO2 eq emissions of different deep-learning models for the classification of renal cell carcinoma on WSIs, also from Rheinisch-Westfälische Technische Hochschule Aachen. We also tested other approaches to reducing CO2 eq emissions, including model pruning and an alternative method for histopathology analysis (pathomics). FINDINGS: The pathology database contained 35 552 specimens (237 179 slides), 6420 of which were prostate specimens (10 115 slides) and 11 801 of which were breast specimens (19 763 slides). We selected and subsequently digitised 140 slides from eight breast-cancer cases and 223 slides from five prostate-cancer cases. Applying large deep-learning models on all WSI tiles of prostate and breast pathology cases would result in yearly CO2 eq emissions of 7·65 metric tons (t; 95% CI 7·62-7·68) with the use of a single deep-learning model per case; yearly CO2 eq emissions were up to 100·56 t (100·21-100·99) with the use of seven deep-learning models per case. CO2 eq emissions for different deep-learning model scenarios, data inputs, and deep-learning model sizes for all slides varied from 3·61 t (3·59-3·63) to 2795·30 t (1177·51-6482·13. For the estimated number of overall pathology cases worldwide, the yearly CO2 eq emissions varied, reaching up to 16 megatons (Mt) of CO2 eq, requiring up to 86 590 km2 (0·22%) of world forest to sequester the CO2 eq emissions. Use of the 7-task scenario and small deep-learning models on slides containing tissue only could substantially reduce CO2 eq emissions worldwide by up to 141 times (0·1 Mt, 95% CI 0·1-0·1). Considering the local environment in Aachen, Germany, the maximum CO2 eq emission from the use of deep learning in digital pathology only would require 32·8% (95% CI 13·8-76·6) of the local forest to sequester the CO2 eq emissions. A single pathomics run on a tissue could provide information that was comparable to or even better than the output of multitask deep-learning models, but with 147 times reduced CO2 eq emissions. INTERPRETATION: Our findings suggest that widespread use of deep learning in pathology might have considerable global-warming potential. The medical community, policy decision makers, and the public should be aware of this potential and encourage the use of CO2 eq emissions reduction strategies where possible. FUNDING: German Research Foundation, European Research Council, German Federal Ministry of Education and Research, Health, Economic Affairs and Climate Action, and the Innovation Fund of the Federal Joint Committee.

Information fusion for fully automated segmentation of head and neck tumors from PET and CT images.

BACKGROUND: PET/CT images combining anatomic and metabolic data provide complementary information that can improve clinical task performance. PET image segmentation algorithms exploiting the multi-modal information available are still lacking. PURPOSE: Our study aimed to assess the performance of PET and CT image fusion for gross tumor volume (GTV) segmentations of head and neck cancers (HNCs) utilizing conventional, deep learning (DL), and output-level voting-based fusions. METHODS: The current study is based on a total of 328 histologically confirmed HNCs from six different centers. The images were automatically cropped to a 200 × 200 head and neck region box, and CT and PET images were normalized for further processing. Eighteen conventional image-level fusions were implemented. In addition, a modified U2-Net architecture as DL fusion model baseline was used. Three different input, layer, and decision-level information fusions were used. Simultaneous truth and performance level estimation (STAPLE) and majority voting to merge different segmentation outputs (from PET and image-level and network-level fusions), that is, output-level information fusion (voting-based fusions) were employed. Different networks were trained in a 2D manner with a batch size of 64. Twenty percent of the dataset with stratification concerning the centers (20% in each center) were used for final result reporting. Different standard segmentation metrics and conventional PET metrics, such as SUV, were calculated. RESULTS: In single modalities, PET had a reasonable performance with a Dice score of 0.77 ± 0.09, while CT did not perform acceptably and reached a Dice score of only 0.38 ± 0.22. Conventional fusion algorithms obtained a Dice score range of [0.76-0.81] with guided-filter-based context enhancement (GFCE) at the low-end, and anisotropic diffusion and Karhunen-Loeve transform fusion (ADF), multi-resolution singular value decomposition (MSVD), and multi-level image decomposition based on latent low-rank representation (MDLatLRR) at the high-end. All DL fusion models achieved Dice scores of 0.80. Output-level voting-based models outperformed all other models, achieving superior results with a Dice score of 0.84 for Majority_ImgFus, Majority_All, and Majority_Fast. A mean error of almost zero was achieved for all fusions using SUVpeak , SUVmean and SUVmedian . CONCLUSION: PET/CT information fusion adds significant value to segmentation tasks, considerably outperforming PET-only and CT-only methods. In addition, both conventional image-level and DL fusions achieve competitive results. Meanwhile, output-level voting-based fusion using majority voting of several algorithms results in statistically significant improvements in the segmentation of HNC.

Effect of Attenuated Light Through Translucent Zirconia on the Interfacial Adaptation and Polymerization of Resin Cements.

PURPOSE: The first objective was to determine if dual-curing of resin cement with reduced light could affect interfacial adaptations of zirconia restoration. The second objective was to examine whether cement type and pretreatment method of universal adhesive affected interfacial adaptation. The final objective was to compare the polymerization degree of cement under different reduced-light conditions. MATERIALS AND METHODS: Inlay cavities were prepared on extracted third molars. Translucent zirconia restorations were milled using Katana UTML (Kuraray Noritake) in three groups with restoration thicknesses of 1, 2, and 3 mm, respectively. Each group had three subgroups using different cementation methods. For subgroup 1, restorations were cemented with self-adhesive cement. For subgroup 2, universal adhesive was applied and light cured. After the restoration was seated with conventional resin cement, light curing was performed. For subgroup 3, after adhesive was applied, the restoration was seated with conventional resin cement. Light curing was performed for the adhesive and cement simultaneously. After thermocycling, interfacial adaptation at the restoration-tooth interface was investigated using swept-source optical coherence tomography imaging. Finally, polymerization shrinkage of the cement was measured using a linometer and compared under the conditions of different zirconia thicknesses and light-curing durations. RESULTS: Interfacial adaptation varied signficantly depending on the zirconia thickness, pretreatment, polymerization mode and cements used (p < 0.05). The effects of the adhesive and polymerization shrinkage differed signficantly, depending on the reduced light under the zirconia (p < 0.05). CONCLUSION: Lower curing-light irradiance may lead to inferior adaptation and lower polymerization of the cement. Polymerization of resin cement can differ depending on the light irradiance and exposure duration.

An interpretable machine learning approach to study the relationship beetwen retrognathia and skull anatomy.

Mandibular retrognathia (C2Rm) is one of the most common oral pathologies. Acquiring a better understanding of the points of impact of C2Rm on the entire skull is of major interest in the diagnosis, treatment, and management of this dysmorphism, but also permits us to contribute to the debate on the changes undergone by the shape of the skull during human evolution. However, conventional methods have some limits in meeting these challenges, insofar as they require defining in advance the structures to be studied, and identifying them using landmarks. In this context, our work aims to answer these questions using AI tools and, in particular, machine learning, with the objective of relaying these treatments automatically. We propose an innovative methodology coupling convolutional neural networks (CNNs) and interpretability algorithms. Applied to a set of radiographs classified into physiological versus pathological categories, our methodology made it possible to: discuss the structures impacted by retrognathia and already identified in literature; identify new structures of potential interest in medical terms; highlight the dynamic evolution of impacted structures according to the level of gravity of C2Rm; provide for insights into the evolution of human anatomy. Results were discussed in terms of the major interest of this approach in the field of orthodontics and, more generally, in the field of automated processing of medical images.

Differential privacy preserved federated transfer learning for multi-institutional 68Ga-PET image artefact detection and disentanglement.

PURPOSE: Image artefacts continue to pose challenges in clinical molecular imaging, resulting in misdiagnoses, additional radiation doses to patients and financial costs. Mismatch and halo artefacts occur frequently in gallium-68 (68Ga)-labelled compounds whole-body PET/CT imaging. Correcting for these artefacts is not straightforward and requires algorithmic developments, given that conventional techniques have failed to address them adequately. In the current study, we employed differential privacy-preserving federated transfer learning (FTL) to manage clinical data sharing and tackle privacy issues for building centre-specific models that detect and correct artefacts present in PET images. METHODS: Altogether, 1413 patients with 68Ga prostate-specific membrane antigen (PSMA)/DOTA-TATE (TOC) PET/CT scans from 3 countries, including 8 different centres, were enrolled in this study. CT-based attenuation and scatter correction (CT-ASC) was used in all centres for quantitative PET reconstruction. Prior to model training, an experienced nuclear medicine physician reviewed all images to ensure the use of high-quality, artefact-free PET images (421 patients' images). A deep neural network (modified U2Net) was trained on 80% of the artefact-free PET images to utilize centre-based (CeBa), centralized (CeZe) and the proposed differential privacy FTL frameworks. Quantitative analysis was performed in 20% of the clean data (with no artefacts) in each centre. A panel of two nuclear medicine physicians conducted qualitative assessment of image quality, diagnostic confidence and image artefacts in 128 patients with artefacts (256 images for CT-ASC and FTL-ASC). RESULTS: The three approaches investigated in this study for 68Ga-PET imaging (CeBa, CeZe and FTL) resulted in a mean absolute error (MAE) of 0.42 ± 0.21 (CI 95%: 0.38 to 0.47), 0.32 ± 0.23 (CI 95%: 0.27 to 0.37) and 0.28 ± 0.15 (CI 95%: 0.25 to 0.31), respectively. Statistical analysis using the Wilcoxon test revealed significant differences between the three approaches, with FTL outperforming CeBa and CeZe (p-value < 0.05) in the clean test set. The qualitative assessment demonstrated that FTL-ASC significantly improved image quality and diagnostic confidence and decreased image artefacts, compared to CT-ASC in 68Ga-PET imaging. In addition, mismatch and halo artefacts were successfully detected and disentangled in the chest, abdomen and pelvic regions in 68Ga-PET imaging. CONCLUSION: The proposed approach benefits from using large datasets from multiple centres while preserving patient privacy. Qualitative assessment by nuclear medicine physicians showed that the proposed model correctly addressed two main challenging artefacts in 68Ga-PET imaging. This technique could be integrated in the clinic for 68Ga-PET imaging artefact detection and disentanglement using multicentric heterogeneous datasets.

Evaluation of physical/mechanical properties of an experimental dental composite modified with a zirconium-based metal-organic framework (MOF) as an innovative dental filler.

OBJECTIVES: This study aimed to modify an experimental dental composite using a synthesized nano-structured methacrylated zirconium-based MOF to enhance physical/mechanical properties. METHODS: The previously known Uio-66-NH2 MOF was first synthesized and post-modified with Glycidyl Methacrylate (GMA). Fourier Transform Infrared (FTIR) Spectroscopy and CHNS analysis confirmed the post-modification reaction. The prepared filler was investigated by XRD, BET, SEM-EDS, and TEM. The experimental composite was prepared by mixing 60% wt. of resin matrix with 40% wt. of fillers, including silanized silica (SS) or Uio-66-NH-Me (UM). The experimental composites' depth of cure (DPC) was investigated in five groups (G1 =40% SS, G2 =30%SS+10%UM, G3 =20%SS+20%UM, G4 =10%SS+30%UM, G5 =40%UM). Then flexural strength(FS), Elastic Modulus(EM), solubility(S), water sorption(WS), degree of conversion(DC), polymerization shrinkage(PS), and polymerization stress(PSR) of the groups with DPC of more than 1 mm were investigated. Finally, the cytotoxicity of composites was studied. RESULTS: The groups with more than 20% wt. UM, filler (G4, G5) had lesser than 1 mm DPC. Therefore, we investigated three groups' physical and mechanical properties with lower than 20% UM filler (G1-G3). Within these groups, G3 has a higher FS, EM (P < 0.05), and lower WS and S (P < 0.05). DC dropped in G2 and G3 compared to G1 (p < 0.05), but there was no significant difference between G2 and G3 (P = 0.594). SIGNIFICANCE: This new filler is an innovative coupling-agent free filler and can be part of dental filler technology itself. It can also introduce a new group of dental fillers based on MOFs, but it still needs a complete investigation to be widely used.

Multi-institutional PET/CT image segmentation using federated deep transformer learning.

BACKGROUND AND OBJECTIVE: Generalizable and trustworthy deep learning models for PET/CT image segmentation necessitates large diverse multi-institutional datasets. However, legal, ethical, and patient privacy issues challenge sharing of datasets between different centers. To overcome these challenges, we developed a federated learning (FL) framework for multi-institutional PET/CT image segmentation. METHODS: A dataset consisting of 328 FL (HN) cancer patients who underwent clinical PET/CT examinations gathered from six different centers was enrolled. A pure transformer network was implemented as fully core segmentation algorithms using dual channel PET/CT images. We evaluated different frameworks (single center-based, centralized baseline, as well as seven different FL algorithms) using 68 PET/CT images (20% of each center data). In particular, the implemented FL algorithms include clipping with the quantile estimator (ClQu), zeroing with the quantile estimator (ZeQu), federated averaging (FedAvg), lossy compression (LoCo), robust aggregation (RoAg), secure aggregation (SeAg), and Gaussian differentially private FedAvg with adaptive quantile clipping (GDP-AQuCl). RESULTS: The Dice coefficient was 0.80±0.11 for both centralized and SeAg FL algorithms. All FL approaches achieved centralized learning model performance with no statistically significant differences. Among the FL algorithms, SeAg and GDP-AQuCl performed better than the other techniques. However, there was no statistically significant difference. All algorithms, except the center-based approach, resulted in relative errors less than 5% for SUVmax and SUVmean for all FL and centralized methods. Centralized and FL algorithms significantly outperformed the single center-based baseline. CONCLUSIONS: The developed FL-based (with centralized method performance) algorithms exhibited promising performance for HN tumor segmentation from PET/CT images.

Finerenone Added to RAS/SGLT2 Blockade for CKD in Alport Syndrome. Results of a Randomized Controlled Trial with Col4a3-/- Mice.

SIGNIFICANCE STATEMENT: We hypothesized that triple therapy with inhibitors of the renin-angiotensin system (RAS), sodium-glucose transporter (SGLT)-2, and the mineralocorticoid receptor (MR) would be superior to dual RAS/SGLT2 blockade in attenuating CKD progression in Col4a3 -deficient mice, a model of Alport syndrome. Late-onset ramipril monotherapy or dual ramipril/empagliflozin therapy attenuated CKD and prolonged overall survival by 2 weeks. Adding the nonsteroidal MR antagonist finerenone extended survival by 4 weeks. Pathomics and RNA sequencing revealed significant protective effects on the tubulointerstitium when adding finerenone to RAS/SGLT2 inhibition. Thus, triple RAS/SGLT2/MR blockade has synergistic effects and might attenuate CKD progression in patients with Alport syndrome and possibly other progressive chronic kidney disorders. BACKGROUND: Dual inhibition of the renin-angiotensin system (RAS) plus sodium-glucose transporter (SGLT)-2 or the mineralocorticoid receptor (MR) demonstrated additive renoprotective effects in large clinical trials. We hypothesized that triple therapy with RAS/SGLT2/MR inhibitors would be superior to dual RAS/SGLT2 blockade in attenuating CKD progression. METHODS: We performed a preclinical randomized controlled trial (PCTE0000266) in Col4a3 -deficient mice with established Alport nephropathy. Treatment was initiated late (age 6 weeks) in mice with elevated serum creatinine and albuminuria and with glomerulosclerosis, interstitial fibrosis, and tubular atrophy. We block-randomized 40 male and 40 female mice to either nil (vehicle) or late-onset food admixes of ramipril monotherapy (10 mg/kg), ramipril plus empagliflozin (30 mg/kg), or ramipril plus empagliflozin plus finerenone (10 mg/kg). Primary end point was mean survival. RESULTS: Mean survival was 63.7±10.0 days (vehicle), 77.3±5.3 days (ramipril), 80.3±11.0 days (dual), and 103.1±20.3 days (triple). Sex did not affect outcome. Histopathology, pathomics, and RNA sequencing revealed that finerenone mainly suppressed the residual interstitial inflammation and fibrosis despite dual RAS/SGLT2 inhibition. CONCLUSION: Experiments in mice suggest that triple RAS/SGLT2/MR blockade may substantially improve renal outcomes in Alport syndrome and possibly other progressive CKDs because of synergistic effects on the glomerular and tubulointerstitial compartments.

Fracture strength and three-dimensional marginal evaluation of biocompatible high-performance polymer versus pressed lithium disilicate crowns.

STATEMENT OF PROBLEM: Despite the acceptable physical properties of biocompatible high-performance polymer (BioHPP), little is known about the marginal accuracy and fracture strength of restorations made from this material. PURPOSE: This in vitro study assessed the marginal and internal adaptation and fracture strength of teeth restored with lithium disilicate (LD) ceramics and BioHPP monolithic crowns. MATERIAL AND METHODS: Twenty-four extracted premolars were prepared for complete coverage crowns and divided into 2 groups to receive pressed IPS e.max LD, or computer-aided design and computer-aided manufacturing (CAD-CAM) BioHPP monolithic crowns. After adhesive cementation, the marginal and internal adaptations of the restorations were evaluated by microcomputed tomography at 18 points for each crown. Specimens were subjected to 6000 thermal cycles at 5 °C and 55 °C and 200 000 load cycles of 100 N at a frequency of 1.2 Hz. The fracture strength of the restorations was then measured in a universal testing machine at a crosshead speed of 0.5 mm/min. Data were analyzed via an independent-sample t-test (α=.05). RESULTS: The mean ±standard deviation of marginal gap was 138.8 ±43.6 µm for LD and 242.1 ±70.7 µm for BioHPP groups (P=.001). The mean ±standard deviation value of absolute marginal discrepancy was 193.8 ±60.8 µm for LD and 263.5 ±97.6 µm for BioHPP groups (P=.06). The internal occlusal and axial gap measurements were 547.5 ±253.1 µm and 197.3 ±54.8 µm for LD (P=.03) and 360 ±62.9 µm and 152.8 ±44.8 µm for BioHPP (P=.04). The mean ±standard deviation of internal space volume was 15.3 ±11.8 µm³ for LD and 24.1 ±10.7 µm³ for BioHPP (P=.08). The mean ±standard deviation of fracture strength was 2509.8 ±680 N for BioHPP and 1090.4 ±454.2 MPa for LD groups (P<.05). CONCLUSIONS: The marginal adaptation of pressed lithium disilicate crowns was better, while BioHPP crowns displayed greater fracture strength. Marginal gap width was not correlated with fracture strength in either group.

Time-dependent structural changes and hypermineralisation of artificially demineralised dentine following treatment with silver diammine fluoride and glass ionomer cement.

OBJECTIVES: This study aimed to evaluate the change of mineral content in dentine lesions over time and examine the effectiveness of the combined treatment with silver diammine fluoride (SDF) and glass ionomer cement (GIC). METHODS: Sixty bovine dentine specimens were divided into 4 groups: cont, Fuji, Safo, and Safo+Fuji. The specimens were imaged and measured using microcomputed tomography (microCT) at 7 time points: pre-demineralisation, after demineralisation for two weeks, immediately after treatment, 1 week, 2 weeks, 1 month, and 3 months after treatment. The 3-month group was evaluated with a light microscope, attenuated total reflectance-Fourier transform infrared (ATR-FTIR) spectroscopy, and scanning electron microscope (SEM)/energy-dispersive X-ray spectroscopy (EDS). Data were analysed by Dunn's test and Wilcoxon signed rank test with Bonferroni correction for microCT, and Kruskal-Wallis test and two-way analysis of variance for EDS characterisation. RESULTS: MicroCT images showed high mineral density beneath dentine lesions in Safo+Fuji. The mineral density at 600 µm in Safo+Fuji increased significantly over time, while Safo showed an opposite trend (adjusted p<0.005). In Safo+Fuji, EDS revealed significantly high energy of fluorine (p<0.05, at 300 µm) and a tendency towards high energy of calcium (p>0.05). However, Safo+Fuji showed lower energy of silver compared to Safo (p<0.001). ATR-FTIR revealed that phosphate groups had the highest peak at a depth between 300 and 400 µm in Safo+Fuji. CONCLUSIONS: Safo+Fuji was effective in remineralising the deep lesion in dentine after one and three months, and a hypermineralisation zone generated beneath the lesion demonstrated additional benefit in this study. CLINICAL SIGNIFICANCE: This long-term in vitro study showed that SDF+GIC treatment could strengthen the structure of decayed teeth when applied in the treatment of patients with advanced rampant caries.

Next-Generation Morphometry for pathomics-data mining in histopathology.

Pathology diagnostics relies on the assessment of morphology by trained experts, which remains subjective and qualitative. Here we developed a framework for large-scale histomorphometry (FLASH) performing deep learning-based semantic segmentation and subsequent large-scale extraction of interpretable, quantitative, morphometric features in non-tumour kidney histology. We use two internal and three external, multi-centre cohorts to analyse over 1000 kidney biopsies and nephrectomies. By associating morphometric features with clinical parameters, we confirm previous concepts and reveal unexpected relations. We show that the extracted features are independent predictors of long-term clinical outcomes in IgA-nephropathy. We introduce single-structure morphometric analysis by applying techniques from single-cell transcriptomics, identifying distinct glomerular populations and morphometric phenotypes along a trajectory of disease progression. Our study provides a concept for Next-generation Morphometry (NGM), enabling comprehensive quantitative pathology data mining, i.e., pathomics.

A flexible framework for anomaly Detection via dimensionality reduction.

Anomaly detection is challenging, especially for large datasets in high dimensions. Here, we explore a general anomaly detection framework based on dimensionality reduction and unsupervised clustering. DRAMA is released as a general python package that implements the general framework with a wide range of built-in options. This approach identifies the primary prototypes in the data with anomalies detected by their large distances from the prototypes, either in the latent space or in the original, high-dimensional space. DRAMA is tested on a wide variety of simulated and real datasets, in up to 3000 dimensions, and is found to be robust and highly competitive with commonly used anomaly detection algorithms, especially in high dimensions. The flexibility of the DRAMA framework allows for significant optimization once some examples of anomalies are available, making it ideal for online anomaly detection, active learning, and highly unbalanced datasets. Besides, DRAMA naturally provides clustering of outliers for subsequent analysis.

Decentralized collaborative multi-institutional PET attenuation and scatter correction using federated deep learning.

PURPOSE: Attenuation correction and scatter compensation (AC/SC) are two main steps toward quantitative PET imaging, which remain challenging in PET-only and PET/MRI systems. These can be effectively tackled via deep learning (DL) methods. However, trustworthy, and generalizable DL models commonly require well-curated, heterogeneous, and large datasets from multiple clinical centers. At the same time, owing to legal/ethical issues and privacy concerns, forming a large collective, centralized dataset poses significant challenges. In this work, we aimed to develop a DL-based model in a multicenter setting without direct sharing of data using federated learning (FL) for AC/SC of PET images. METHODS: Non-attenuation/scatter corrected and CT-based attenuation/scatter corrected (CT-ASC) 18F-FDG PET images of 300 patients were enrolled in this study. The dataset consisted of 6 different centers, each with 50 patients, with scanner, image acquisition, and reconstruction protocols varying across the centers. CT-based ASC PET images served as the standard reference. All images were reviewed to include high-quality and artifact-free PET images. Both corrected and uncorrected PET images were converted to standardized uptake values (SUVs). We used a modified nested U-Net utilizing residual U-block in a U-shape architecture. We evaluated two FL models, namely sequential (FL-SQ) and parallel (FL-PL) and compared their performance with the baseline centralized (CZ) learning model wherein the data were pooled to one server, as well as center-based (CB) models where for each center the model was built and evaluated separately. Data from each center were divided to contribute to training (30 patients), validation (10 patients), and test sets (10 patients). Final evaluations and reports were performed on 60 patients (10 patients from each center). RESULTS: In terms of percent SUV absolute relative error (ARE%), both FL-SQ (CI:12.21-14.81%) and FL-PL (CI:11.82-13.84%) models demonstrated excellent agreement with the centralized framework (CI:10.32-12.00%), while FL-based algorithms improved model performance by over 11% compared to CB training strategy (CI: 22.34-26.10%). Furthermore, the Mann-Whitney test between different strategies revealed no significant differences between CZ and FL-based algorithms (p-value > 0.05) in center-categorized mode. At the same time, a significant difference was observed between the different training approaches on the overall dataset (p-value < 0.05). In addition, voxel-wise comparison, with respect to reference CT-ASC, exhibited similar performance for images predicted by CZ (R2 = 0.94), FL-SQ (R2 = 0.93), and FL-PL (R2 = 0.92), while CB model achieved a far lower coefficient of determination (R2 = 0.74). Despite the strong correlations between CZ and FL-based methods compared to reference CT-ASC, a slight underestimation of predicted voxel values was observed. CONCLUSION: Deep learning-based models provide promising results toward quantitative PET image reconstruction. Specifically, we developed two FL models and compared their performance with center-based and centralized models. The proposed FL-based models achieved higher performance compared to center-based models, comparable with centralized models. Our work provided strong empirical evidence that the FL framework can fully benefit from the generalizability and robustness of DL models used for AC/SC in PET, while obviating the need for the direct sharing of datasets between clinical imaging centers.

Deep Learning-based Non-rigid Image Registration for High-dose Rate Brachytherapy in Inter-fraction Cervical Cancer.

In this study, an inter-fraction organ deformation simulation framework for the locally advanced cervical cancer (LACC), which considers the anatomical flexibility, rigidity, and motion within an image deformation, was proposed. Data included 57 CT scans (7202 2D slices) of patients with LACC randomly divided into the train (n = 42) and test (n = 15) datasets. In addition to CT images and the corresponding RT structure (bladder, cervix, and rectum), the bone was segmented, and the coaches were eliminated. The correlated stochastic field was simulated using the same size as the target image (used for deformation) to produce the general random deformation. The deformation field was optimized to have a maximum amplitude in the rectum region, a moderate amplitude in the bladder region, and an amplitude as minimum as possible within bony structures. The DIRNet is a convolutional neural network that consists of convolutional regressors, spatial transformation, as well as resampling blocks. It was implemented by different parameters. Mean Dice indices of 0.89 ± 0.02, 0.96 ± 0.01, and 0.93 ± 0.02 were obtained for the cervix, bladder, and rectum (defined as at-risk organs), respectively. Furthermore, a mean average symmetric surface distance of 1.61 ± 0.46 mm for the cervix, 1.17 ± 0.15 mm for the bladder, and 1.06 ± 0.42 mm for the rectum were achieved. In addition, a mean Jaccard of 0.86 ± 0.04 for the cervix, 0.93 ± 0.01 for the bladder, and 0.88 ± 0.04 for the rectum were observed on the test dataset (15 subjects). Deep learning-based non-rigid image registration is, therefore, proposed for the high-dose-rate brachytherapy in inter-fraction cervical cancer since it outperformed conventional algorithms.

Recommendations on compiling test datasets for evaluating artificial intelligence solutions in pathology.

Artificial intelligence (AI) solutions that automatically extract information from digital histology images have shown great promise for improving pathological diagnosis. Prior to routine use, it is important to evaluate their predictive performance and obtain regulatory approval. This assessment requires appropriate test datasets. However, compiling such datasets is challenging and specific recommendations are missing. A committee of various stakeholders, including commercial AI developers, pathologists, and researchers, discussed key aspects and conducted extensive literature reviews on test datasets in pathology. Here, we summarize the results and derive general recommendations on compiling test datasets. We address several questions: Which and how many images are needed? How to deal with low-prevalence subsets? How can potential bias be detected? How should datasets be reported? What are the regulatory requirements in different countries? The recommendations are intended to help AI developers demonstrate the utility of their products and to help pathologists and regulatory agencies verify reported performance measures. Further research is needed to formulate criteria for sufficiently representative test datasets so that AI solutions can operate with less user intervention and better support diagnostic workflows in the future.

The effect of different finishing polishing protocols on stain absorption and color stability of resin composite restorations.

PURPOSE: To evaluate the degree of stain absorption of resin composite following different finishing and polishing protocols and to evaluate the efficacy of prophylaxis paste with or without addition of carbamide peroxide to remove composite surface stain. METHODS: 50 resin composite blocks were prepared using nano-hybrid composite (Filtek Z250XT) and giomer (Beautifil II LS). Specimens were either polished with a disk or coated with 5th/7th generation bonding or glycerin gel. The color parameters (lightness, chroma and hue) of the specimens were measured prior to staining, and after 1 and 2 weeks of staining using a color chromometer. The composite surface stain removal capability of prophylaxis pastes with or without the addition of carbamide peroxide was evaluated before and after polishing. RESULTS: The repeated measures ANOVA showed that composite polished with a polishing disk or coated with glycerin gel have better color stability compared to composite without any polishing or coated with 5th or 7th generation bonding materials. The prophylaxis paste was able to remove composite surface stain; however, the addition of carbamide peroxide did not show any additional benefit. The results suggested that polishing with polishing disks or coating with glycerin gel is useful for the color stability of resin composite. CLINICAL SIGNIFICANCE: Selection of finishing and polishing protocols varies among individual practitioners. Each of these protocols has some pros and cons concerning the esthetic and color stability of composite. The present results suggested that the use of polishing disk or glycerin gel during polymerization would be a useful protocol to achieve and maintain esthetically stable composite restorations.