Classification of OCT images for the early diagnosis of Alzheimer's disease

Sandeep, C S; Kumar, Atul; Mahadevan, Kittappa M.; Manoj, P K

doi:10.1109/i2c2.2017.8321780

Cited by 5 publications

(6 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, assigning the correct value to the activity depends on the availability of data from the company about previous customers for a fairly long period of time. When manually counting leads, points can be assigned based on a set of fixed rules [7,[18][19][20]. The table in Figure 1 is formed as follows: the first column Activity contains any user action.…”

Section: A Manual Evaluation Of Potential Customersmentioning

confidence: 99%

“…Data from previous periods were used as training data for classification algorithms, and data on purchasing moments were used as limits for contacts who made a purchase in the past. Different ways of aggregating data over time have been considered to ensure that customer constraints do not affect the final model [2,15,19]. The performance of the model was evaluated using cross-validation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Model for Clustering and Predicting Customer Behavior in a Structured Data Set

Boyko¹

2023

JEFMS

View full text Add to dashboard Cite

The main goal of the work is to study the process of leads coring. In this paper, two ways of lead generation are explained – manual and using predictive models. This condones, per investigation, is the most efficient way to score a lead.The analytical review starts with an overview of existing lead scoring methods, describing each and comparing the pros and cons. There is an analytical comparison of the products that already exist on the market, such as HubSpot, Infer, PipeCandy and Maroon.ai. As a conclusion from this comparison, the effect of this study has a benefit in that it is a low-weight plugin with a significantly lower price than the market median. To provide a list of methods and algorithms used and describe why each was selected for a specific goal. Four aggregations were made to choose the best-fitting model, resulting in the LightGBM selection. There is a comprehensive description of the LighGBM features and math background, along with an explanation of GOSS. Also, after the model was selected, the set data manipulations were done, such as aggregation, class weight balance, tuning and exhaustive analysis, and correction of disbalance. In part to the results, there is an extensive analysis of the model performance after studying the data set and usage in a real-case CRM – Salesforce. As results showed – the created plugin can be easily integrated into any CRM solution using the native marketplaces or package-delivery systems.

show abstract

Section: A Manual Evaluation Of Potential Customersmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Model for Clustering and Predicting Customer Behavior in a Structured Data Set

Boyko¹

2023

JEFMS

View full text Add to dashboard Cite

show abstract

“…If several types have the same number of occurrences in k nearest neighbours, the class with the smaller sequence number is assigned. Then the classified object is added to the general sample and considered in classifying the following ones [17,19], [28].…”

Section: Formulation Of the Problem Actuality Of The Classification P...mentioning

confidence: 99%

“…Then, among these k objects, we need to count how often this class occurs and assign the type that occurs most often to our entity. If two categories' occurrences are the largest and the same, then the class with the smaller number is assigned [7,19], [25].…”

Section: Introductionmentioning

confidence: 99%

Methodsof analysis of multimodal data to increase the accuracy of classification

Boyko¹,

Muzyka²

2022

AAIT

View full text Add to dashboard Cite

This paper proposes methods for analyzing multimodal data that will help improve the overall accuracy of the results and plans for classifying K-Nearest Neighbor (KNN) to minimize their risk. The mechanism of increasing the accuracy of KNN classification is considered. The research methods used in this work are comparison, analysis, induction, and experiment. This work aimed to improve the accuracy of KNN classification by comparing existing algorithms and applying new methods. Many literary and media sources on the classification according to the algorithm k of the nearest neighbors were analyzed, and the most exciting variations of the given algorithm were selected. Emphasis will be placed on achieving maximum classification accuracy by comparing existing and improving methods for choosing the number k and finding the nearest class. Algorithms with and without data analysis and preprocessing are also compared. All the strategies discussed in this article will be achieved purely practically. An experimental classification by k nearest neighbors with different variations was performed. Data for the experiment used two different data sets of various sizes. Different classifications k and the test sample size were taken as classification arguments. The paper studies three variants of the algorithm k nearest neighbors: the classical KNN, KNN with the lowest average and hybrid KNN. These algorithms are compared for different test sample sizes for other numbers k. The article analyzes the data before classification. As for selecting the number k, no simple method would give the maximum result with great accuracy. The essence of the algorithm is to find k closest to the sample of objects already classified by predefined and numbered classes. Then, among these k objects, you need to count how often the class occurs and assign the most common class to the selected object. If two classes' occurrences are the largest and the same, the class with the smaller number is assigned

show abstract

“…Especially for MRI data, a preprocessing was performed to extract features, such as total intracranial volume, hippocampal volume, and entorhinal cortex thickness, which are relevant to predicting MCI conversion. Recent methods (Lama et al, 2017; Sandeep et al, 2017) that extract features also can be used before running our package. The summary statistics of samples and hyperparameters are shown in Table 1 .…”

Section: Classification Task On the Real Datasetmentioning

confidence: 99%

MildInt: Deep Learning-Based Multimodal Longitudinal Data Integration Framework

et al. 2019

View full text Add to dashboard Cite

As large amounts of heterogeneous biomedical data become available, numerous methods for integrating such datasets have been developed to extract complementary knowledge from multiple domains of sources. Recently, a deep learning approach has shown promising results in a variety of research areas. However, applying the deep learning approach requires expertise for constructing a deep architecture that can take multimodal longitudinal data. Thus, in this paper, a deep learning-based python package for data integration is developed. The python package deep learning-based m ult i modal l ongitudinal d ata int egration framework (MildInt) provides the preconstructed deep learning architecture for a classification task. MildInt contains two learning phases: learning feature representation from each modality of data and training a classifier for the final decision. Adopting deep architecture in the first phase leads to learning more task-relevant feature representation than a linear model. In the second phase, linear regression classifier is used for detecting and investigating biomarkers from multimodal data. Thus, by combining the linear model and the deep learning model, higher accuracy and better interpretability can be achieved. We validated the performance of our package using simulation data and real data. For the real data, as a pilot study, we used clinical and multimodal neuroimaging datasets in Alzheimer’s disease to predict the disease progression. MildInt is capable of integrating multiple forms of numerical data including time series and non-time series data for extracting complementary features from the multimodal dataset.

show abstract

Classification of OCT images for the early diagnosis of Alzheimer's disease

Cited by 5 publications

References 23 publications

A Model for Clustering and Predicting Customer Behavior in a Structured Data Set

A Model for Clustering and Predicting Customer Behavior in a Structured Data Set

Methodsof analysis of multimodal data to increase the accuracy of classification

MildInt: Deep Learning-Based Multimodal Longitudinal Data Integration Framework

Contact Info

Product

Resources

About