Tag: statistics

Indian J Tuberc. 2023 Apr;70(2):168-175. doi: 10.1016/j.ijtb.2022.03.028. Epub 2022 Apr 6.

ABSTRACT

BACKGROUND: TB continues to ravage high burden countries despite aggressive TB control measures. Poverty and adverse socioeconomic and cultural factors play a significant role in stigmatization, causing delayed health care seeking, non-compliance to treatment and spread of disease in the community. Women are more vulnerable to stigmatization, posing the risk of gender inequality in health care. The objectives of this study were to ascertain the degree of stigmatization and gender disparity in TB related stigma in the community.

METHODS: Study was conducted among TB unaffected persons, using consecutive sampling from bystanders of patients attending the hospital for diseases other than TB. Closed structured questionnaire was used for measuring socio-demographic, knowledge and stigma variables. Stigma scoring was done using TB vignette.

RESULTS: Majority subjects (119 males and 102 females) were from rural area and low socioeconomic status; more than 60% of males and females having college education. Half the subjects answered more than half the TB knowledge questions correctly. Knowledge score was significantly lower among females compared with males (p < 0.002) despite high literacy. Overall stigma scoring was low (mean score = 15.9; total 75). Stigma was higher among females compared with males (p < 0.002); more profound among females receiving female vignettes (Chi-square = 14.1, p < 0.0001). The association was significant even after adjusting for co-variables (OR = 3.323, P = 0.005). Low knowledge showed minimal (statistically insignificant) association with stigma.

CONCLUSIONS: Perceived stigma though low, was more among females and much higher with female vignette, indicating significant gender disparity in stigma towards TB.

PMID:37100573 | DOI:10.1016/j.ijtb.2022.03.028

Artif Intell Med. 2023 May;139:102542. doi: 10.1016/j.artmed.2023.102542. Epub 2023 Apr 5.

ABSTRACT

BACKGROUND/INTRODUCTION: Manual detection and localization of the brain’s epileptogenic areas using electroencephalogram (EEG) signals is time-intensive and error-prone. An automated detection system is, thus, highly desirable for support in clinical diagnosis. A set of relevant and significant non-linear features plays a major role in developing a reliable, automated focal detection system.

METHODS: A new feature extraction method is designed to classify focal EEG signals using eleven non-linear geometrical attributes derived from the Fourier-Bessel series expansion-based empirical wavelet transform (FBSE-EWT) segmented rhythm’s second-order difference plot (SODP). A total of 132 features (2 channels × 6 rhythms × 11 geometrical attributes) were computed. However, some of the obtained features might be non-significant and redundant features. Hence, to acquire an optimal set of relevant non-linear features, a new hybridization of ‘Kruskal-Wallis statistical test (KWS)’ with ‘VlseKriterijuska Optimizacija I Komoromisno Resenje’ termed as the KWS-VIKOR approach was adopted. The KWS-VIKOR has a two-fold operational feature. First, the significant features are selected using the KWS test with a p-value lesser than 0.05. Next, the multi-attribute decision-making (MADM) based VIKOR method ranks the selected features. Several classification methods further validate the efficacy of the features of the selected top n%.

RESULTS: The proposed framework has been evaluated using the Bern-Barcelona dataset. The highest classification accuracy of 98.7% was achieved using the top 35% ranked features in classifying the focal and non-focal EEG signals with the least-squares support vector machine (LS-SVM) classifier.

CONCLUSIONS: The achieved results exceeded those reported through other methods. Hence, the proposed framework will more effectively assist the clinician in localizing the epileptogenic areas.

PMID:37100511 | DOI:10.1016/j.artmed.2023.102542

Artif Intell Med. 2023 May;139:102539. doi: 10.1016/j.artmed.2023.102539. Epub 2023 Mar 31.

ABSTRACT

Certain life-threatening abnormalities, such as cholangiocarcinoma, in the human biliary tract are curable if detected at an early stage, and ultrasonography has been proven to be an effective tool for identifying them. However, the diagnosis often requires a second opinion from experienced radiologists, who are usually overwhelmed by many cases. Therefore, we propose a deep convolutional neural network model, named biliary tract network (BiTNet), developed to solve problems in the current screening system and to avoid overconfidence issues of traditional deep convolutional neural networks. Additionally, we present an ultrasound image dataset for the human biliary tract and demonstrate two artificial intelligence (AI) applications: auto-prescreening and assisting tools. The proposed model is the first AI model to automatically screen and diagnose upper-abdominal abnormalities from ultrasound images in real-world healthcare scenarios. Our experiments suggest that prediction probability has an impact on both applications, and our modifications to EfficientNet solve the overconfidence problem, thereby improving the performance of both applications and of healthcare professionals. The proposed BiTNet can reduce the workload of radiologists by 35% while keeping the false negatives to as low as 1 out of every 455 images. Our experiments involving 11 healthcare professionals with four different levels of experience reveal that BiTNet improves the diagnostic performance of participants of all levels. The mean accuracy and precision of the participants with BiTNet as an assisting tool (0.74 and 0.61, respectively) are statistically higher than those of participants without the assisting tool (0.50 and 0.46, respectively (p<0.001)). These experimental results demonstrate the high potential of BiTNet for use in clinical settings.

PMID:37100509 | DOI:10.1016/j.artmed.2023.102539

Artif Intell Med. 2023 May;139:102536. doi: 10.1016/j.artmed.2023.102536. Epub 2023 Mar 29.

ABSTRACT

OBJECTIVE: Many Computer Aided Prognostic (CAP) systems based on machine learning techniques have been proposed in the field of oncology. The objective of this systematic review was to assess and critically appraise the methodologies and approaches used in predicting the prognosis of gynecological cancers using CAPs.

METHODS: Electronic databases were used to systematically search for studies utilizing machine learning methods in gynecological cancers. Study risk of bias (ROB) and applicability were assessed using the PROBAST tool. 139 studies met the inclusion criteria, of which 71 predicted outcomes for ovarian cancer patients, 41 predicted outcomes for cervical cancer patients, 28 predicted outcomes for uterine cancer patients, and 2 predicted outcomes for gynecological malignancies broadly.

RESULTS: Random forest (22.30 %) and support vector machine (21.58 %) classifiers were used most commonly. Use of clinicopathological, genomic and radiomic data as predictors was observed in 48.20 %, 51.08 % and 17.27 % of studies, respectively, with some studies using multiple modalities. 21.58 % of studies were externally validated. Twenty-three individual studies compared ML and non-ML methods. Study quality was highly variable and methodologies, statistical reporting and outcome measures were inconsistent, preventing generalized commentary or meta-analysis of performance outcomes.

CONCLUSION: There is significant variability in model development when prognosticating gynecological malignancies with respect to variable selection, machine learning (ML) methods and endpoint selection. This heterogeneity prevents meta-analysis and conclusions regarding the superiority of ML methods. Furthermore, PROBAST-mediated ROB and applicability analysis demonstrates concern for the translatability of existing models. This review identifies ways that this can be improved upon in future works to develop robust, clinically translatable models within this promising field.

PMID:37100507 | DOI:10.1016/j.artmed.2023.102536

Artif Intell Med. 2023 May;139:102534. doi: 10.1016/j.artmed.2023.102534. Epub 2023 Mar 28.

ABSTRACT

BACKGROUND: Indigenous peoples often have higher rates of morbidity and mortality associated with cardiometabolic disease (CMD) than non-Indigenous people and this may be even more so in urban areas. The use of electronic health records and expansion of computing power has led to mainstream use of artificial intelligence (AI) to predict the onset of disease in primary health care (PHC) settings. However, it is unknown if AI and in particular machine learning is used for risk prediction of CMD in Indigenous peoples.

METHODS: We searched peer-reviewed literature using terms associated with AI machine learning, PHC, CMD, and Indigenous peoples.

RESULTS: We identified 13 suitable studies for inclusion in this review. Median total number of participants was 19,270 (range 911-2,994,837). The most common algorithms used in machine learning in this setting were support vector machine, random forest, and decision tree learning. Twelve studies used the area under the receiver operating characteristic curve (AUC) to measure performance. Two studies reported an AUC of >0.9. Six studies had an AUC score between 0.9 and 0.8, 4 studies had an AUC score between 0.8 and 0.7. 1 study reported an AUC score between 0.7 and 0.6. Risk of bias was observed in 10 (77 %) studies.

CONCLUSION: AI machine learning and risk prediction models show moderate to excellent discriminatory ability over traditional statistical models in predicting CMD. This technology could help address the needs of urban Indigenous peoples by predicting CMD early and more rapidly than conventional methods.

PMID:37100506 | DOI:10.1016/j.artmed.2023.102534

Tob Control. 2023 Apr 26:tc-2023-057931. doi: 10.1136/tc-2023-057931. Online ahead of print.

ABSTRACT

BACKGROUND: In 2020, Ethiopia passed a landmark tax proclamation implementing an evidence-based mixed excise system aimed at curbing tobacco use. This study evaluates the impact of the tax increase of more than 600% on both legal and illegal cigarette prices in order to gauge the impact of the tax reform in the presence of a sizeable illicit cigarette market.

METHODS: Data on 1774 cigarette prices were obtained from retailers during Empty Cigarette Pack Surveys in the capital and major regional cities conducted in 2018 and 2022. Packs were categorised as ‘legal’ or ‘illicit’ using criteria from the tobacco control directives. Descriptive and regression analyses were used to study the cigarette price changes during the period of 2018-2022, capturing the impact of the 2020 tax increase.

RESULT: Prices of both legal and illegal cigarettes increased in response to the tax increase. In 2018, the stick prices ranged from ETB0.88 (Ethiopian birr) to ETB5.00 for legal cigarettes while they ranged from ETB0.75 to ETB3.25 for illegal ones. In 2022, a legal stick sold for ETB01.50-ETB2.73 and an illegal stick for ETB1.92-ETB8.00. The average real price of legal and illegal brands increased by 18% and 37%, respectively. The multivariate analysis confirms that prices of illicit cigarettes grew faster compared with the legal ones. By 2022, illicit brands were on average more expensive compared with their legal counterparts. This result is statistically significant at p<0.01.

CONCLUSION: The prices of both legal and illegal cigarettes increased following the 2020 tax increase, with the average real cigarette price increasing by 24%. As a result, the tax increase likely had a positive impact on public health despite a sizeable illicit cigarette market.

PMID:37100452 | DOI:10.1136/tc-2023-057931

Acta Psychiatr Scand. 2023 Apr 26. doi: 10.1111/acps.13561. Online ahead of print.

ABSTRACT

INTRODUCTION: Use of antipsychotic drugs, especially second-generation agents, has been suggested to cause acute pancreatitis in multiple case reports; however, such an association has not been corroborated by larger studies. This study examined the association of antipsychotic drugs with risk of acute pancreatitis.

METHODS: Nationwide case-control study, based on data from several Swedish registers and including all 52,006 cases of acute pancreatitis diagnosed in Sweden between 2006 and 2019 (with up to 10 controls per case; n = 518,081). Conditional logistic regression models were used to calculate odds ratios (ORs) in current and past users of first-generation and second-generation antipsychotic drugs (dispensed prescription <91 and ≥91 days of the index date, respectively) compared with never users of such drugs.

RESULTS: In the crude model, first-generation and second-generation antipsychotic drugs were associated with increased risk of acute pancreatitis, with slightly higher ORs for past use (1.58 [95% confidence interval 1.48-1.69] and 1.39 [1.29-1.49], respectively) than for current use (1.34 [1.21-1.48] and 1.24 [1.15-1.34], respectively). The ORs were largely attenuated in the multivariable model-which included, among others, alcohol abuse and the Charlson comorbidity index-up to the point where only a statistically significant association remained for past use of first-generation agents (OR 1.18 [1.10-1.26]).

CONCLUSION: There was no clear association between use of antipsychotic drugs and risk of acute pancreatitis in this very large case-control study, indicating that previous case report data are most likely explained by confounding.

PMID:37100434 | DOI:10.1111/acps.13561

Acta Psychiatr Scand. 2023 Apr 26. doi: 10.1111/acps.13556. Online ahead of print.

NO ABSTRACT

PMID:37100433 | DOI:10.1111/acps.13556

Int J Sports Physiol Perform. 2023 Apr 26:1-8. doi: 10.1123/ijspp.2022-0458. Online ahead of print.

ABSTRACT

BACKGROUND: The development and influence of physical capabilities and game action performance over the course of the season are a big challenge for coaches and players.

PURPOSE: The aims of the present study were to examine (1) the seasonal changes in the physical capabilities (mechanical and kinematic) and game-performance indicators in top-level men volleyball players and (2) the relationship between these physical capabilities and game-performance indicators in official matches.

METHODS: Eleven top-level players participated. Players were physically tested 3 times during the season. Before each test, players’ match performance (11 sets) was analyzed according to the level of opposition and match location. The percentage of change, statistical differences over the season (Friedman and Wilcoxon tests), and associations between variables (Spearman r) were calculated (P < .05) among mechanical (force-velocity profile during vertical jump and bench press), kinematic (jump height and spike ball speed), and game action performance features (coefficient, efficacy, and percentage of errors in serve, attack, and block).

RESULTS: The theoretical maximal force and velocity during vertical jump and bench press, respectively; the peak spike ball speed; and the serve efficacy significantly increased over the season. Moreover, there was a significant reduction in serve errors as the jump height increased (r = -.44; P = .026), as well as a significant increase in serve errors as the peak spike ball speed increased (r = -.62; P = .001).

CONCLUSION: These findings reveal how the physical and game action performance variables evolve and interact during the season. This may help coaches and trainers to monitor and analyze the most relevant volleyball performance factors.

PMID:37100426 | DOI:10.1123/ijspp.2022-0458

Global Spine J. 2023 Apr 26:21925682231172431. doi: 10.1177/21925682231172431. Online ahead of print.

NO ABSTRACT

PMID:37100407 | DOI:10.1177/21925682231172431