Machine Learning Assisted Signal Analysis in Acoustic Microscopy for Non-Destructive Defect Identification

2019 ◽  
Author(s):  
Michael Kögel ◽  
Sebastian Brand ◽  
Frank Altmann
Keyword(s):  
Machine Learning ◽  
Failure Analysis ◽  
Human Error ◽  
Flip Chip ◽  
Acoustic Microscopy ◽  
Classification Model ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
Scanning Acoustic Microscopy ◽  
Chip Technology

Abstract Signal processing and data interpretation in scanning acoustic microscopy is often challenging and based on the subjective decisions of the operator, making the defect classification results prone to human error. The aim of this work was to combine unsupervised and supervised machine learning techniques for feature extraction and image segmentation that allows automated classification and predictive failure analysis on scanning acoustic microscopy (SAM) data. In the first part, conspicuous signal components of the time-domain echo signals and their weighting matrices are extracted using independent component analysis. The applicability was shown by the assisted separation of signal patterns to intact and defective bumps from a dataset of a CPU-device manufactured in flip-chip technology. The high success-rate was verified by physical cross-sectioning and high-resolution imaging. In the second part, the before mentioned signal separation was employed to generate a labeled dataset for training and finetuning of a classification model based on a one-dimensional convolutional neural network. The learning model was sensitive to critical features of the given task without human intervention for classification between intact bumps, defective bumps and background. This approach was evaluated on two individual test samples that contained multiple defects in the solder bumps and has been verified by physical inspection. The verification of the classification model reached an accuracy of more than 97% and was successfully applied to an unknown sample which demonstrates the high potential of machine learning concepts for further developments in assisted failure analysis.

Failure Analysis of Flip-Chip Interconnections Through Acoustic Microscopy

1996 ◽  
Author(s):  
O. Diaz de Leon ◽  
M. Nassirian ◽  
C. Todd ◽  
R. Chowdhury
Keyword(s):  
Failure Analysis ◽  
Large Scale ◽  
Flip Chip ◽  
Ultrasonic Waves ◽  
Acoustic Microscopy ◽  
Acoustic Microscope ◽  
Chip Technology ◽  
Ball Joints ◽  
Non Destructive ◽  
Scanning Acoustic Microscope

Abstract Integration of circuits on semiconductor devices with resulting increase in pin counts is driving the need for improvements in packaging for functionality and reliability. One solution to this demand is the Flip- Chip concept in Ultra Large Scale Integration (ULSI) applications [1]. The flip-chip technology is based on the direct attach principle of die to substrate interconnection.. The absence of bondwires clearly enables packages to become more slim and compact, and also provides higher pin counts and higher-speeds [2]. However, due to its construction, with inherent hidden structures the Flip-Chip technology presents a challenge for non-destructive Failure Analysis (F/A). The scanning acoustic microscope (SAM) has recently emerged as a valuable evaluation tool for this purpose [3]. C-mode scanning acoustic microscope (C-SAM), has the ability to demonstrate non-destructive package analysis while imaging the internal features of this package. Ultrasonic waves are very sensitive, particularly when they encounter density variations at surfaces, e.g. variations such as voids or delaminations similar to air gaps. These two anomalies are common to flip-chips. The primary issue with this package technology is the non-uniformity of the die attach through solder ball joints and epoxy underfill. The ball joints also present defects as open contacts, voids or cracks. In our acoustic microscopy study packages with known defects are considered. It includes C-SCAN analysis giving top views at a particular package interface and a B-SCAN analysis that provides cross-sectional views at a desired point of interest. The cross-section analysis capability gives confidence to the failure analyst in obtaining information from a failing area without physically sectioning the sample and destroying its electrical integrity. Our results presented here prove that appropriate selection of acoustic scanning modes and frequency parameters leads to good reliable correlation between the physical defects in the devices and the information given by the acoustic microscope.

scholarly journals Prediction of Spam Email using Machine Learning Classification Algorithm

2021 ◽  
Vol 9 (VI) ◽  
pp. 1107-1112
Author(s):  
P Sai Teja
Keyword(s):  
Machine Learning ◽  
Performance Metrics ◽  
Classification Model ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
K Nearest Neighbor ◽  
Adaptive Boosting ◽  
Machine Learning Classification ◽  
E Mail

Unsolicited e-mail also known as Spam has become a huge concern for each e-mail user. In recent times, it is very difficult to filter spam emails as these emails are produced or created or written in a very special manner so that anti-spam filters cannot detect such emails. This paper compares and reviews performance metrics of certain categories of supervised machine learning techniques such as SVM (Support Vector Machine), Random Forest, Decision Tree, CNN, (Convolutional Neural Network), KNN(K Nearest Neighbor), MLP(Multi-Layer Perceptron), Adaboost (Adaptive Boosting) Naïve Bayes algorithm to predict or classify into spam emails. The objective of this study is to consider the details or content of the emails, learn a finite dataset available and to develop a classification model that will be able to predict or classify whether an e-mail is spam or not.

Thin-Die Flip Chip Physical-FA Process Flow

2002 ◽  
Author(s):  
Andrew J. Komrowski ◽  
N. S. Somcio ◽  
Daniel J. D. Sullivan ◽  
Charles R. Silvis ◽  
Luis Curiel ◽  
...  
Keyword(s):  
Sample Preparation ◽  
Failure Analysis ◽  
Flip Chip ◽  
Semiconductor Industry ◽  
Organic Substrate ◽  
Fault Isolation ◽  
Preparation Methods ◽  
Chip Technology ◽  
Isolation Test ◽  
Sample Preparation Methods

Abstract The use of flip chip technology inside component packaging, so called flip chip in package (FCIP), is an increasingly common package type in the semiconductor industry because of high pin-counts, performance and reliability. Sample preparation methods and flows which enable physical failure analysis (PFA) of FCIP are thus in demand to characterize defects in die with these package types. As interconnect metallization schemes become more dense and complex, access to the backside silicon of a functional device also becomes important for fault isolation test purposes. To address these requirements, a detailed PFA flow is described which chronicles the sample preparation methods necessary to isolate a physical defect in the die of an organic-substrate FCIP.

Extending Acoustic Microscopy for Comprehensive Failure Analysis Applications

2010 ◽  
Author(s):  
Sebastian Brand ◽  
Matthias Petzold ◽  
Peter Czurratis ◽  
Peter Hoffrogge
Keyword(s):  
Image Processing ◽  
Failure Analysis ◽  
Signal Analysis ◽  
Flip Chip ◽  
Fault Isolation ◽  
Acoustic Microscopy ◽  
Specific Information ◽  
Full Potential ◽  
Time Frequency ◽  
Non Destructive

Abstract In industrial manufacturing of microelectronic components, non-destructive failure analysis methods are required for either quality control or for providing a rapid fault isolation and defect localization prior to detailed investigations requiring target preparation. Scanning acoustic microscopy (SAM) is a powerful tool enabling the inspection of internal structures in optically opaque materials non-destructively. In addition, depth specific information can be employed for two- and three-dimensional internal imaging without the need of time consuming tomographic scan procedures. The resolution achievable by acoustic microscopy is depending on parameters of both the test equipment and the sample under investigation. However, if applying acoustic microscopy for pure intensity imaging most of its potential remains unused. The aim of the current work was the development of a comprehensive analysis toolbox for extending the application of SAM by employing its full potential. Thus, typical case examples representing different fields of application were considered ranging from high density interconnect flip-chip devices over wafer-bonded components to solder tape connectors of a photovoltaic (PV) solar panel. The progress achieved during this work can be split into three categories: Signal Analysis and Parametric Imaging (SA-PI), Signal Analysis and Defect Evaluation (SA-DE) and Image Processing and Resolution Enhancement (IP-RE). Data acquisition was performed using a commercially available scanning acoustic microscope equipped with several ultrasonic transducers covering the frequency range from 15 MHz to 175 MHz. The acoustic data recorded were subjected to sophisticated algorithms operating in time-, frequency- and spatial domain for performing signal- and image analysis. In all three of the presented applications acoustic microscopy combined with signal- and image processing algorithms proved to be a powerful tool for non-destructive inspection.

Failure Analysis of Broken Stitch Bonds in TSSOP Packages Using Scanning Acoustic Microscopy (SAM) and Time Domain Reflectometry (TDR)

2004 ◽  
Author(s):  
Bilal Abd-AlRahman ◽  
Corey Lewis ◽  
Todd Simons
Keyword(s):  
Failure Analysis ◽  
Mechanical Stress ◽  
Time Domain ◽  
Open Circuit ◽  
Time Domain Reflectometry ◽  
Acoustic Microscopy ◽  
Scanning Acoustic Microscopy ◽  
Root Cause ◽  
Analysis Application

Abstract A failure analysis application utilizing scanning acoustic microscopy (SAM) and time domain reflectometry (TDR) for failure analysis has been developed to isolate broken stitch bonds in thin shrink small outline package (TSSOP) devices. Open circuit failures have occurred in this package due to excessive bending of the leads during assembly. The tools and their specific application to this technique as well as the limitations of C-SAM, TDR and radiographic analyses are discussed. By coupling C-SAM and TDR, a failure analyst can confidently determine whether the cause of an open circuit in a TSSOP package is located at the stitch bond. The root cause of the failure was determined to be abnormal mechanical stress placed on the pins during the lead forming operation. While C-SAM and TDR had proven useful in the analysis of TSSOP packages, it can potentially be expanded to other wire-bonded packages.

Failure and Stress Analysis of Cu TSVs Using GHz-Scanning Acoustic Microscopy and Scanning Infrared Polariscopy

2015 ◽  
Author(s):  
Ingrid De Wolf ◽  
Ahmad Khaled ◽  
Martin Herms ◽  
Matthias Wagner ◽  
Tatjana Djuric ◽  
...  
Keyword(s):  
Failure Analysis ◽  
Stress Analysis ◽  
Rayleigh Waves ◽  
Acoustic Microscopy ◽  
Scanning Acoustic Microscopy ◽  
Through Silicon Vias ◽  
Stress Anisotropy ◽  
Silicon Vias ◽  
Ic Technology ◽  
Detection Of Defects

Abstract This paper discusses the application of two different techniques for failure analysis of Cu through-silicon vias (TSVs), used in 3D stacked-IC technology. The first technique is GHz Scanning Acoustic Microscopy (GHz- SAM), which not only allows detection of defects like voids, cracks and delamination, but also the visualization of Rayleigh waves. GHz-SAM can provide information on voids, delamination and possibly stress near the TSVs. The second is a reflection-based photoelastic technique (SIREX), which is shown to be very sensitive to stress anisotropy in the Si near TSVs and as such also to any defect affecting this stress, such as delamination and large voids.

Application of Machine Learning Techniques to Predict Binding Affinity for Drug Targets: A Study of Cyclin-Dependent Kinase 2

2020 ◽  
Vol 28 (2) ◽  
pp. 253-265 ◽  
Author(s):  
Gabriela Bitencourt-Ferreira ◽  
Amauri Duarte da Silva ◽  
Walter Filgueira de Azevedo
Keyword(s):  
Machine Learning ◽  
Binding Affinity ◽  
Predictive Performance ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
Scoring Functions ◽  
Cyclin Dependent Kinase ◽  
Learning Models ◽  
Learning Techniques ◽  
Machine Learning Models

Background: The elucidation of the structure of cyclin-dependent kinase 2 (CDK2) made it possible to develop targeted scoring functions for virtual screening aimed to identify new inhibitors for this enzyme. CDK2 is a protein target for the development of drugs intended to modulate cellcycle progression and control. Such drugs have potential anticancer activities. Objective: Our goal here is to review recent applications of machine learning methods to predict ligand- binding affinity for protein targets. To assess the predictive performance of classical scoring functions and targeted scoring functions, we focused our analysis on CDK2 structures. Methods: We have experimental structural data for hundreds of binary complexes of CDK2 with different ligands, many of them with inhibition constant information. We investigate here computational methods to calculate the binding affinity of CDK2 through classical scoring functions and machine- learning models. Results: Analysis of the predictive performance of classical scoring functions available in docking programs such as Molegro Virtual Docker, AutoDock4, and Autodock Vina indicated that these methods failed to predict binding affinity with significant correlation with experimental data. Targeted scoring functions developed through supervised machine learning techniques showed a significant correlation with experimental data. Conclusion: Here, we described the application of supervised machine learning techniques to generate a scoring function to predict binding affinity. Machine learning models showed superior predictive performance when compared with classical scoring functions. Analysis of the computational models obtained through machine learning could capture essential structural features responsible for binding affinity against CDK2.

scholarly journals Predictive Modelling of Employee Turnover in Indian IT Industry Using Machine Learning Techniques

2019 ◽  
Vol 23 (1) ◽  
pp. 12-21 ◽  
Author(s):  
Shikha N. Khera ◽  
Divya
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Confusion Matrix ◽  
Predictive Modelling ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
It Industry ◽  
Knowledge Based ◽  
Employee Attrition

Information technology (IT) industry in India has been facing a systemic issue of high attrition in the past few years, resulting in monetary and knowledge-based loses to the companies. The aim of this research is to develop a model to predict employee attrition and provide the organizations opportunities to address any issue and improve retention. Predictive model was developed based on supervised machine learning algorithm, support vector machine (SVM). Archival employee data (consisting of 22 input features) were collected from Human Resource databases of three IT companies in India, including their employment status (response variable) at the time of collection. Accuracy results from the confusion matrix for the SVM model showed that the model has an accuracy of 85 per cent. Also, results show that the model performs better in predicting who will leave the firm as compared to predicting who will not leave the company.

scholarly journals Local mortality estimates during the COVID-19 pandemic in Italy

2021 ◽  
Author(s):  
Augusto Cerqua ◽  
Roberta Di Stefano ◽  
Marco Letta ◽  
Sara Miccoli
Keyword(s):  
Machine Learning ◽  
Excess Mortality ◽  
Control Method ◽  
Local Level ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
Mortality Data ◽  
Official Method ◽  
Learning Techniques ◽  
Mortality Estimates

AbstractEstimates of the real death toll of the COVID-19 pandemic have proven to be problematic in many countries, Italy being no exception. Mortality estimates at the local level are even more uncertain as they require stringent conditions, such as granularity and accuracy of the data at hand, which are rarely met. The “official” approach adopted by public institutions to estimate the “excess mortality” during the pandemic draws on a comparison between observed all-cause mortality data for 2020 and averages of mortality figures in the past years for the same period. In this paper, we apply the recently developed machine learning control method to build a more realistic counterfactual scenario of mortality in the absence of COVID-19. We demonstrate that supervised machine learning techniques outperform the official method by substantially improving the prediction accuracy of the local mortality in “ordinary” years, especially in small- and medium-sized municipalities. We then apply the best-performing algorithms to derive estimates of local excess mortality for the period between February and September 2020. Such estimates allow us to provide insights about the demographic evolution of the first wave of the pandemic throughout the country. To help improve diagnostic and monitoring efforts, our dataset is freely available to the research community.

Sign in / Sign up

Export Citation Format

Share Document