minimum sample
Recently Published Documents


TOTAL DOCUMENTS

174
(FIVE YEARS 46)

H-INDEX

22
(FIVE YEARS 3)

Blood ◽  
2021 ◽  
Vol 138 (Supplement 1) ◽  
pp. 3056-3056
Author(s):  
Kareem Midlig ◽  
Regina Draliuk ◽  
Meir Preis ◽  
Barbara Silverman ◽  
Mouna Ballan

Abstract Background : Multiple myeloma (MM) is one of the most common hematological malignancies. The disease is characterized by multiple symptoms resulting from the disease itself, from complications related to therapy, and as a result of the involvement of other organ systems. MM influences various aspects of patient's and family's lives. Therefore, there is a need to better understand the balance between disease control and symptoms management. Objectives : The main goal of this study is to emphasize the power and importance of Patient Reported Outcome Measures (PROMs) and Family Reported Outcome Measures (FROMs) as additional tools for patient assessment. This study evaluated the correlation between Patient Reported Outcome Measures (PROMs) and Family Reported Outcome Measures (FROMs) and disease evaluation according to the International Myeloma Working Group (IMW) response criteria in active myeloma patients. A comparison between patient and family reporting (PROMs & FROMs) and the staging of the disease according to the revised international staging system (R-ISS) was done. In addition, this study examined the confounders that may explain the relationship between PROMs and FROMs and disease evaluation. Methods : This is a quantitative, prospective, observational and longitudinal study of active patients with MM. After receiving Carmel Institutional Review Board approval to conduct the study, we enrolled fifty seven MM patients, the participants completed questionnaires of PROMs and FROMs at intervals of 3 months for one year. In addition, we monitored multiple clinical measures of patient response to treatment. A descriptive analysis of the research variables has been performed; differences between the PROMs/FROM and clinical variables analyzed by Pearson correlation, comparing PROMs/FROMs mean at the beginning of the study with the results at 3, 6, 9 and 12 months . A mixed regression model was used to examine the predictive ability of the study. In other words, the ability of Patient/Family Reported Outcome to predict the disease evaluation. Sample size was calculated using Win-Pepi software, using 5% significance and 80% power. For a coefficient of 0.4 between Patient and Family Reported Outcome and MM clinical evaluations, the minimum sample size required is 47 patients, for a coefficient of 0.35, the minimum sample size required is 62 patients. for a coefficient of 0.50, the minimum sample size required is 37 patients. This study recruited a sample size of 57 patients. Results - Fifty-seven patients participated in this study. After 3 months of treatment, a better disease evaluation was associated with improvement in disease symptoms or side effects reported by the patient. Furthermore, a better disease response was associated with a better body image scale and better future perspective. We observed a similar association after 6 and 9 months. In addition, the more the patient reported side effects or disease symptoms, the more it affects the family member (PROMs were positively correlated with FROMs). A better body image and future perspective reported by patient was associated with a lower effect on family member (PROMs were negatively correlated with FROMs)). A positive significant correlation was found between physician ranking of physical status ECOG (Eastern Cooperative Oncology Group) and the effect on family members. In other words, the worse the physical status of the patient, the more it affect the family member (in months 0,3,6 and 9). These finding were supported by the mixed model analysis, which showed a significant effect of disease symptoms, appetite loss, physical function, future perspective, and global satisfaction in prediction of clinical status. Conclusion- There is a significant relation between PROM/FROM and the typical assessment tools. This study highlights the power of PROM/FROM tools to evaluate patient from his point of view and to adjust the treatment accordingly. Finally, this study raises up the importance of continuing the research about the effect on the family member as a result of the patient disease and clinical status. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.


2021 ◽  
Author(s):  
Daniel M. Goldenholz ◽  
Haoqi Sun ◽  
Wolfgang Ganglberger ◽  
M. Brandon Westover

ABSTRACTOBJECTIVEBefore integrating new machine learning (ML) into clinical practice, algorithms must undergo validation. Validation studies require sample size estimates. Unlike hypothesis testing studies seeking a p-value, the goal of validating predictive models is obtaining estimates of model performance. Our aim was to provide a standardized, data distribution- and model-agnostic approach to sample size calculations for validation studies of predictive ML models.MATERIALS AND METHODSSample Size Analysis for Machine Learning (SSAML) was tested in three previously published models: brain age to predict mortality (Cox Proportional Hazard), COVID hospitalization risk prediction (ordinal regression), and seizure risk forecasting (deep learning). The SSAML steps are: 1) Specify performance metrics for model discrimination and calibration. For discrimination, we use area under the receiver operating curve (AUC) for classification and Harrell’s C-statistic for survival models. For calibration, we employ calibration slope and calibration-in-the-large. 2) Specify the required precision and accuracy (≤0.5 normalized confidence interval width and ±5% accuracy). 3) Specify the required coverage probability (95%). 4) For increasing sample sizes, calculate the expected precision and bias that is achievable. 5) Choose the minimum sample size that meets all requirements.RESULTSMinimum sample sizes were obtained in each dataset using standardized criteria.DISCUSSIONSSAML provides a formal expectation of precision and accuracy at a desired confidence level.CONCLUSIONSSAML is open-source and agnostics to data type and ML model. It can be used for clinical validation studies of ML models.


2021 ◽  
pp. 41-79
Author(s):  
Sylwester Białowąs ◽  
Blaženka Knežević ◽  
Iwona Olejnik ◽  
Magdalena Stefańska

The main goal of the chapter is to present the basics of survey research that can be used in analyzes of sustainable development. The first part presents the measurement levels. The basic characteristic of every variable is its level of measurement. It implies the following analysis and available techniques. This part introduces four levels of measurements: nominal, ordinal, interval and ratio, showing their characteristics and examples. Then the focus is on the implications of a given level of measurement on the possibilities of the statistical analysis. The aim of the second chapter is to explain the foundations of preparing a questionnaire for the research on the issues related to sustainable development. An example of an organic food questionnaire is also provided. The third part presents considerations necessary for the sampling process. The main goal is to present the basic methods of calculating the minimum sample size, as well as the methods of its selection. This section presents the arguments for conducting the study on a sample rather than on the entire population, and also several formulas enabling the calculation of the minimum sample size. A discussion of the most important methods of selecting respondents to the sample—both random and non-random, can also be found here. The last two parts of this chapter, describe the ways of presenting the results of quantitative research. They describe first view of the variables including frequency distribution with charts, central tendency measures and cross-tabulation. Finally, the methods of presenting research results obtained on the basis of the Likert scale and other examples of data visualization schemes are presented.


2021 ◽  
pp. 096228022110463
Author(s):  
Glen P Martin ◽  
Richard D Riley ◽  
Gary S Collins ◽  
Matthew Sperrin

Recent minimum sample size formula (Riley et al.) for developing clinical prediction models help ensure that development datasets are of sufficient size to minimise overfitting. While these criteria are known to avoid excessive overfitting on average, the extent of variability in overfitting at recommended sample sizes is unknown. We investigated this through a simulation study and empirical example to develop logistic regression clinical prediction models using unpenalised maximum likelihood estimation, and various post-estimation shrinkage or penalisation methods. While the mean calibration slope was close to the ideal value of one for all methods, penalisation further reduced the level of overfitting, on average, compared to unpenalised methods. This came at the cost of higher variability in predictive performance for penalisation methods in external data. We recommend that penalisation methods are used in data that meet, or surpass, minimum sample size requirements to further mitigate overfitting, and that the variability in predictive performance and any tuning parameters should always be examined as part of the model development process, since this provides additional information over average (optimism-adjusted) performance alone. Lower variability would give reassurance that the developed clinical prediction model will perform well in new individuals from the same population as was used for model development.


Author(s):  
A. B. Zoramawa ◽  
S. U. Gulumbe

This paper proposed a sequential probability sampling plan for a truncated life test using a Rayleigh distribution from  a designed double sampling plans where the interest was to obtain the minimum sample size necessary to assure that the average life time of a product is longer than the default life time at the specified consumer’s and producer’s confidence level. Estimations of minimum sample, acceptance and rejection numbers obtained are analyzed and presented to explain the usefulness of sequential plans in relation to single and double sampling plan. Probability of acceptance (Pa), Average sample number (ASN) and Average outgoing quality (AOQ) for the plans are computed. The three regions; acceptance, continue sampling and rejection were determined. The five points necessary to plot ASN curve were also computed and presented.


2021 ◽  
Vol 11 (19) ◽  
pp. 8881
Author(s):  
Esther Kho ◽  
Behdad Dashtbozorg ◽  
Joyce Sanders ◽  
Marie-Jeanne T. F. D. Vrancken Peeters ◽  
Frederieke van Duijnhoven ◽  
...  

Developing algorithms for analyzing hyperspectral images as an intraoperative tool for margin assessment during breast-conserving surgery requires a dataset with reliable histopathologic labels. The feasibility of using tissue slices hyperspectral dataset with a high correlation with histopathology for developing an algorithm for analyzing the images from the surface of lumpectomy specimens was investigated. We presented a method to acquire hyperspectral images from the lumpectomy surface with a high correlation with histopathology. The tissue slices dataset was compared with the dataset obtained on lumpectomy specimen and the wavelengths with a penetration depth up to the minimum sample thickness of the tissue slices were used to develop a tissue classification algorithm. Spectral differences were observed between tissue slices and lumpectomy datasets due to differences in the sample thickness between both datasets; wavelengths with a high penetration depth were able to penetrate through the thinner tissue slices, affecting the captured signal. By using only wavelengths with a penetration depth up to the minimum sample thickness of the tissue slices, the adipose tissue could be discriminated from other tissue types, but differentiating malignant from connective tissue was more challenging.


2021 ◽  
Vol 7 (2) ◽  
Author(s):  
Miriam Zemanova

Research on animals is one of the most controversial ethical issues in our society. It is imperative that animal welfare is being considered and the harm and distress to animals used in research is minimized. This could be achieved through implementation of the so-called 3Rs principles for animal research, which are now implemented in many legislations worldwide. These principles serve as a basis for research without the use of animals (Replacement), with as few animals as possible (Reduction), and in which the animal’s welfare is as good as possible (Refinement). While there has been a lot of focus on implementation of these principles, only a few studies have documented the knowledge and adoption of the 3Rs among researchers. One field that has been particularly neglected is ecological research, which can involve many practices that affect animal welfare. Moreover, the knowledge, experience, and attitudes about animal use in ecological research and education has never been examined before. In order to close this gap, I conducted a survey among European ecologists. Responses from 107 respondents from 23 countries revealed that lethal and invasive research methods are prevalent, and that more than half of the respondents have never heard of the 3Rs principles for animal research. Major concerns are also the lack of calculation of the minimum sample size and widespread of dissection classes as part of education. Additionally, most respondents experienced ethical doubts about their research, and did not receive any training in animal welfare or ethics. These findings revealed that it is necessary to implement rigorous standards for ecological research and enforce the implementation of the 3Rs principles. Furthermore, the evaluation of current educational practices in ecology is urgently needed.


Sign in / Sign up

Export Citation Format

Share Document