Tag: statistics

EJNMMI Phys. 2022 Jul 19;9(1):47. doi: 10.1186/s40658-022-00476-w.

ABSTRACT

BACKGROUND: In recent years, a lot of effort has been put in the enhancement of medical imaging using artificial intelligence. However, limited patient data in combination with the unavailability of a ground truth often pose a challenge to a systematic validation of such methodologies. The goal of this work was to investigate a recently proposed method for an artificial intelligence-based generation of synthetic SPECT projections, for acceleration of the image acquisition process based on a large dataset of realistic SPECT simulations.

METHODS: A database of 10,000 SPECT projection datasets of heterogeneous activity distributions of randomly placed random shapes was simulated for a clinical SPECT/CT system using the SIMIND Monte Carlo program. Synthetic projections at fixed angular increments from a set of input projections at evenly distributed angles were generated by different u-shaped convolutional neural networks (u-nets). These u-nets differed in noise realization used for the training data, number of input projections, projection angle increment, and number of training/validation datasets. Synthetic projections were generated for 500 test projection datasets for each u-net, and a quantitative analysis was performed using statistical hypothesis tests based on structural similarity index measure and normalized root-mean-squared error. Additional simulations with varying detector orbits were performed on a subset of the dataset to study the effect of the detector orbit on the performance of the methodology. For verification of the results, the u-nets were applied to Jaszczak and NEMA physical phantom data obtained on a clinical SPECT/CT system.

RESULTS: No statistically significant differences were observed between u-nets trained with different noise realizations. In contrast, a statistically significant deterioration was found for training with a small subset (400 datasets) of the 10,000 simulated projection datasets in comparison with using a large subset (9500 datasets) for training. A good agreement between synthetic (i.e., u-net generated) and simulated projections before adding noise demonstrates a denoising effect. Finally, the physical phantom measurements show that our findings also apply for projections measured on a clinical SPECT/CT system.

CONCLUSION: Our study shows the large potential of u-nets for accelerating SPECT/CT imaging. In addition, our analysis numerically reveals a denoising effect when generating synthetic projections with a u-net. Clinically interesting, the methodology has proven robust against camera orbit deviations in a clinically realistic range. Lastly, we found that a small number of training samples (e.g., ~ 400 datasets) may not be sufficient for reliable generalization of the u-net.

PMID:35852673 | DOI:10.1186/s40658-022-00476-w

Arch Gynecol Obstet. 2022 Jul 19. doi: 10.1007/s00404-022-06685-z. Online ahead of print.

ABSTRACT

INTRODUCTION: The management of fetal macrosomia remains controversial. A protocol for induction of labor in the case of a suspected macrosomic fetus has been in place in our maternity hospital since 2016. We studied the impact of this protocol by analyzing the mode of delivery. We then studied its safety in terms of maternal and fetal morbidity and mortality and the risk factors of macrosomia.

METHODS: Retrospective, comparative, single-center study including 256 patients between 2016 and 2020 in a type 3 maternity hospital in France. We compared 114 patients induced at 39 weeks of gestation (fetal weight ≥ 95th p; group 1) with 142 patients who after 37 weeks of gestation delivered a macrosomic newborn (≥ 95th p according to Audipog; group 2) not diagnosed antenatally.

RESULTS: The rate of vaginal delivery in the group 1 was 78.9% vs 83.8% in group 2 (p = 0.318). The rate of neonatal acidosis in group 1 was statistically lower than in group 2 (5.2% vs 16.9%; p = 0.004). The other maternal and neonatal outcomes were not significantly different in the two groups. A previous macrosomic newborn appeared to be a risk factor for macrosomia (p = 0.02).

CONCLUSION: The establishment of a protocol for the induction of labor in the case of macrosomia in our maternity hospital did not increase the rate of vaginal delivery, but has a neonatal benefit, by significantly reducing neonatal acidosis.

PMID:35852647 | DOI:10.1007/s00404-022-06685-z

Med Oncol. 2022 Jul 19;39(10):152. doi: 10.1007/s12032-022-01741-9.

ABSTRACT

Excision repair complementary complex 5 (ERCC5) is an important component in the repair pathway of platinum-induced damage. The current study evaluated the effect of ERCC5 variants (rs751402 and rs1047768) on the clinical outcome of platinum-based regimens in non-small cell lung cancer (NSCLC) patients. A prospective, cohort study was conducted on 57 newly diagnosed NSCLC Egyptian patients. Patients received either cisplatin or carboplatin-based chemotherapy. DNA was extracted and the variants were analyzed using real time PCR. This study found no significant difference between the studied variants and patients’ response to chemotherapy, progression-free survival (PFS) or overall survival (OS). However, a statistically significant association was found between the histologic subtypes and the studied variants (p = 0.028 and 0.018 for rs751402 and rs1047768, respectively). A statistically significant association was evident between the type of the allele present in the studied polymorphisms, p value = 0.000040. Moreover, the minor allele frequency (MAF) of the studied variants rs751402 and rs1047768 were similar to those of African and European populations, respectively. Results of this study have concluded that ERCC5 variants did not affect the clinical outcome of platinum-based chemotherapy in NSCLC. A significant coinheritance was found between the two variants of ERCC5. Moreover, the similarity between the MAF of the studied variants and the African or European population can guide future research when extrapolating data from African European populations to their Egyptian counterparts.

PMID:35852645 | DOI:10.1007/s12032-022-01741-9

Acta Neurochir (Wien). 2022 Jul 19. doi: 10.1007/s00701-022-05282-y. Online ahead of print.

ABSTRACT

BACKGROUND: The aim of this study was to investigate whether clinical outcomes in patients aged ≥ 70 undergoing decompressive surgery for degenerative cervical myelopathy (DCM) differ from those of younger patients (50-70 years) at 1 year.

METHODS: Data were obtained from the Norwegian Registry for Spine Surgery (NORspine). Among 651 patients included, 177 (27.2%) were ≥ 70 years old. The primary outcome was change in the Neck Disability Index (NDI). Secondary outcomes were changes in the European Myelopathy Score (EMS), quality of life (EuroQoL EQ-5D), numeric rating scales (NRS) for headache, neck pain, and arm pain, and complications.

RESULTS: Significant improvements in all patient-reported outcomes (PROMs) were detected for both age cohorts at 1 year. For the two age cohorts combined, there was a statistically significant improvement in the NDI score (mean 9.2, 95% CI 7.7 to 10.6, P < 0.001). There were no differences between age cohorts in mean change of NDI (- 8.9 vs. – 10.1, P = 0.48), EQ-5D (0.13 vs. 0.17, P = 0.37), or NRS pain scores, but elderly patients experienced a larger improvement in EMS (0.7 vs. 1.3, P = 0.02). A total of 74 patients (15.6%) in the younger cohort and 43 patients (24.3%) in the older cohort experienced complications or adverse effects within 3 months of surgery, mainly urinary and respiratory tract infections.

CONCLUSION: Surgery for DCM was associated with significant improvement across a wide range of PROMs for both younger and elderly patients. Surgery for DCM should not be denied based on age alone.

PMID:35852626 | DOI:10.1007/s00701-022-05282-y

J Cancer Res Clin Oncol. 2022 Jul 19. doi: 10.1007/s00432-022-04201-z. Online ahead of print.

ABSTRACT

PURPOSE: Immune checkpoint inhibitors (ICIs) have improved the prognosis of cancer patients significantly with few predictive makers for treatment efficiency. Since interferon-gamma (IFN-γ) displayed its association with immunotherapy, we explored the correlation between IFN-γ and the efficacy of ICIs in tumor treatment.

METHODS: We retrospectively examined cancer patients who received immune checkpoint inhibitors as first-line therapy at the Fourth Hospital of Hebei Medical University. The patients were divided into a low concentration group of IFN-γ (≤ 1.2 pg/mL) and a high concentration group (≥ 1.3 pg/mL) to evaluate the efficacy, which was indicated by the objective response rate (ORR), disease control rate (DCR), progression-free survival (PFS) and overall survival (OS).

RESULTS: Thirty-five patients with low IFN-γ and 56 patients with high IFN-γ were involved in the evaluation, and the DCR was significantly different between these two groups (p = 0.009) with a high group of 81.4% (95% CI 69-94%) and a low group of 51.9% (95% CI 32-72%). The subsequent Kaplan-Meier survival analysis showed that the high IFN-γ patients displayed longer median OS than that of the low IFN-γ patients (p = 0.049), while no statistical difference existed for PFS (p = 0.971). The multivariate analysis also confirmed that the high IFN-γ level was independently associated with a better prognosis (HR: 0.318 95% CI 0.113-0.894, p = 0.030).

CONCLUSIONS: Basal serum IFN-γ levels were associated with the DCR and OS of cancer patients with higher IFN-γ exhibiting beneficial efficiency for ICIs treatment.

PMID:35852620 | DOI:10.1007/s00432-022-04201-z

Z Gerontol Geriatr. 2022 Jul 19. doi: 10.1007/s00391-022-02086-8. Online ahead of print.

ABSTRACT

BACKGROUND: The intervention is a multiday health program of the Social Insurance for Agriculture, Forestry and Horticulture (SVLFG) for informal caregivers.

AIM OF THE WORK: Does the intervention increase the use of outside help to gain free time for self-care?

MATERIAL AND METHODS: Using a quasi-experimental within-design, we analyze annual benefit data from the SVLFG long-term care insurance for the cluster sample Bavaria from 2017 to 2020 with intervention in 2018/2019. Using fixed effects panel regressions, we determine the effect heterogeneity for care relationship, duration of care and degree of care, adjusting for period effects (intervention and comparison group: 88 and 6045 persons with 207 and 16,091 observations, respectively).

RESULT: After the intervention, the use of services for outside help tends to be higher than before. The effect is significantly more pronounced with increasing care level, care within the partnership and shorter care duration. With care within the partnership, start of care before 1 year and care level 4/5, more outside help is used afterwards for 1160 € (p < 0.001) than before.

DISCUSSION: Design and statistical method would allow a causal conclusion; however, in view of the limitations, in particular the small number of cases in the intervention group and the only limited characteristics available to control for time-dynamic heterogeneity, caution is advisable.

PMID:35852589 | DOI:10.1007/s00391-022-02086-8

Epilepsy Behav. 2022 Jul 15;134:108840. doi: 10.1016/j.yebeh.2022.108840. Online ahead of print.

ABSTRACT

AIM: This study was conducted to examine the knowledge and attitudes that students in the field of healthcare services have about epilepsy.

METHOD: This cross-sectional and descriptive study was conducted with 312 students of the Department of Health Care Services (home patient and elderly care) in Turkey. The data were collected using a sociodemographic information form, the Epilepsy Knowledge Level Scale, and the Attitudes Towards Epilepsy Scale. Descriptive statistics, the Shapiro-Wilk test, the Mann-Whitney U test, the Kruskal-Wallis H test and the Spearman correlation test were used to analyze the data. Significance was considered as p < 0.05.

RESULTS: Students in the field of healthcare services constituted the sample of the study, and more than half of the students (68.3%) were female. The mean scores obtained from the Epilepsy Knowledge Scale and the Epilepsy Attitude Scale were 8.38 ± 4.29 and 60.02 ± 7.28 respectively. A positive and significant relationship was found between the students’ knowledge and attitude scores about epilepsy (p < 0.01). It was found that students’ attitudes toward epilepsy positively increased as their knowledge increased. It was observed that female students who received education about epilepsy, witnessed seizures, and had relatives with epilepsy had more positive attitudes toward epilepsy.

CONCLUSION: In this study, it was found that the student participants had a moderate level of knowledge about epilepsy and exhibited a positive attitude toward epilepsy. It was determined that the students’ attitudes toward epilepsy positively increased as their level of knowledge about epilepsy increased.

PMID:35849866 | DOI:10.1016/j.yebeh.2022.108840

Theriogenology. 2022 Jul 11;190:1-7. doi: 10.1016/j.theriogenology.2022.07.002. Online ahead of print.

ABSTRACT

The aim of this study was to evaluate the postoperative analgesic and antioxidant effects of butorphanol given in the preoperative or early postoperative period. Twenty-seven healthy female dogs undergoing ovariohysterectomy were randomly divided into three groups as before surgery group (BSG, n = 7) received butorphanol 30 min before preanesthetic administration, after surgery group (ASG, n = 10) received butorphanol during the last skin suture and the control group (CG, n = 10) received no butorphanol. Pain was assessed with short form of the Glasgow composite pain scale (CMPS-SF). Serum concentration of malondialdehyde (MDA), superoxide dismutase (SOD) and glutathione peroxidase activities (GPx) were quantified by spectrophotometric methods to assess oxidative stress status. The pain score increased rapidly at 1 h after surgery and then decreased gradually towards to 24 h in all groups. There was no statistical difference among the groups in terms of CMPS-SF scores (P > 0.05). Serum concentration of MDA was lower in ASG than in BSG and CG from 1 h to 24 h after surgery. Serum activity of GPx was higher in ASG than in BSG and CG from 2 h to 24 h (P < 0.05). Serum activity of SOD was higher in ASG than in BSG and CG from 1 h to 24 h after surgery (P < 0.05). Serum SOD activity at different time points in ASG did not differ compared to preoperative level though it decreased significantly from 1 h onwards both in CT and BSG. The results indicate that single butorphanol administration either before or after the operation might not provide sufficient analgesia, however, it seems that it has antioxidant potential and may protect tissues by reducing oxidative stress when administered early postoperative period following ovariohysterectomy.

PMID:35849850 | DOI:10.1016/j.theriogenology.2022.07.002

JMIR Form Res. 2022 Jul 14. doi: 10.2196/37059. Online ahead of print.

ABSTRACT

BACKGROUND: To reduce person-to-person contact, the COVID-19 pandemic has driven a massive shift to virtual care. Defined as the use of technology (synchronous or asynchronous) to support communication between healthcare providers and patients, rural-urban differences in virtual are relatively unexplored.

OBJECTIVE: The two-fold purpose of the present study was to examine rural and urban virtual care access, use, and satisfaction during the pandemic and to identify any unmet needs.

METHODS: A cross-sectional online survey exploring virtual care among rural and urban adults in summer 2021 using a combination of fixed and open-ended response options. Quantitative data were analyzed using both descriptive and inferential statistics, and qualitative data were analyzed using inductive thematic content analysis.

RESULTS: Overall, 501 (373, 74.4% female; Age 19-86 years; 237, 47.3% rural-living) Western Canadians completed the survey. Virtual care use was high among both rural (171/237, 72.2%) and urban (188/264, 71.2%) participants, with over half reporting having only started to use virtual care since the pandemic. The self-reported need for mental health programs and services increased during the pandemic compared to prior for both rural and urban participants. Among virtual care users, interest in its continuation was high. Our analysis also shows that internet quality (all Ps<.05) and eHealth literacy (all Ps<.001) were positively associated with participants’ perceptions of virtual care usefulness, ease of use, and satisfaction with no rural/urban differences. Rural participants were less likely to have used video in communicating with doctors/healthcare providers, compared to urban participants (P<.001). When describing unmet needs, participants described: 1) lack of access to care; 2) limited health promotion and prevention options; and 3) lack of mental health service options.

CONCLUSIONS: The increased demand for and use of virtual care may reflect increased availability and a lack of alternatives due to limited in-person services during the COVID-19 pandemic, so a balance between virtual care and in-person care is important to consider post-pandemic. Further, ensuring availability of high-speed internet and education to support patients will be important for providing accessible and effective virtual care, especially for rural residents.

PMID:35849794 | DOI:10.2196/37059

J Chem Inf Model. 2022 Jul 18. doi: 10.1021/acs.jcim.1c01334. Online ahead of print.

ABSTRACT

The field of machine learning for drug discovery is witnessing an explosion of novel methods. These methods are often benchmarked on simple physicochemical properties such as solubility or general druglikeness, which can be readily computed. However, these properties are poor representatives of objective functions in drug design, mainly because they do not depend on the candidate compound’s interaction with the target. By contrast, molecular docking is a widely applied method in drug discovery to estimate binding affinities. However, docking studies require a significant amount of domain knowledge to set up correctly, which hampers adoption. Here, we present dockstring, a bundle for meaningful and robust comparison of ML models using docking scores. dockstring consists of three components: (1) an open-source Python package for straightforward computation of docking scores, (2) an extensive dataset of docking scores and poses of more than 260,000 molecules for 58 medically relevant targets, and (3) a set of pharmaceutically relevant benchmark tasks such as virtual screening or de novo design of selective kinase inhibitors. The Python package implements a robust ligand and target preparation protocol that allows nonexperts to obtain meaningful docking scores. Our dataset is the first to include docking poses, as well as the first of its size that is a full matrix, thus facilitating experiments in multiobjective optimization and transfer learning. Overall, our results indicate that docking scores are a more realistic evaluation objective than simple physicochemical properties, yielding benchmark tasks that are more challenging and more closely related to real problems in drug discovery.

PMID:35849793 | DOI:10.1021/acs.jcim.1c01334