Halil Altay Güvenir
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Halil Altay Güvenir has a total of 85 co-authors. Here are the top co-authors by number of shared publications:
Faruk Polat
Middle East Technical University
4 shared publicationsMurat Kurtcephe
Computer Engineering, Bilkent University, Ankara, 06800, Turkey
4 shared publicationsDeniz Katircioglu-Öztürk
Medical Information Technology Solutions (MITS), Bilkent University Cyberpark, Ankara, Turkey
3 shared publicationsEmre Oto
Medical Information Technology Solutions (MITS), Bilkent University Cyberpark, Ankara, Turkey
3 shared publications- Ahmet Rüçhan Akar
Department of Cardiovascular Surgery, Heart Center, Ankara University Faculty of Medicine, Cebeci Hospital, Ankara, Türkiye
2 shared publications Ali Oto
Department of Cardiology, Memorial Ankara Hospital, Ankara, Turkey
2 shared publications
Latest Publications
- Open Access
- 20 May 2026
- Published by MDPI AG in Machine Learning and Knowledge Extraction
- Vol. 8 (5) , 137
Generating Multiple-Choice Knowledge Questions with Interpretable Difficulty Estimation Using Knowledge Graphs and Large Language Models
Generating multiple-choice questions (MCQs) with difficulty estimation remains challenging in automated MCQ-generation systems used in adaptive, AI-assisted education. This study proposes a novel methodology for generating MCQs with difficulty estimation from input documents by utilizing knowledge graphs (KGs) and large language models (LLMs). Our approach uses an LLM to construct a KG from input documents, from which MCQs are then systematically generated. Each MCQ is generated by selecting a node from the KG as the key, sampling a related triple or quintuple—optionally augmented with an extra triple—and prompting an LLM to generate a corresponding stem from these graph components. Distractors are then selected from the KG. For each MCQ, nine difficulty signals are computed and combined into a unified difficulty score using a data-driven approach. Within a 150-MCQ, proof-of-concept dataset from Wikipedia, the proposed signals show interpretable associations with empirical incorrect-answer rates aligning with human responses/performance. The results support the feasibility of the proposed pipeline, yet a larger-scale human study may be required to establish deployment-scale validity. Our approach improves automated MCQ generation by integrating structured knowledge representations with LLMs and a data-driven difficulty estimation model.
- 20 May 2026
- Published by MDPI AG in Machine Learning and Knowledge Extraction
- Vol. 8 (5) , 137
Open AccessGenerating Multiple-Choice Knowledge Questions with Interpretable Difficulty Estimation Using Knowledge Graphs and Large Language Models
Generating multiple-choice questions (MCQs) with difficulty estimation remains challenging in automated MCQ-generation systems used in adaptive, AI-assisted education. This study proposes a novel methodology for generating MCQs with difficulty estimation from input documents by utilizing knowledge graphs (KGs) and large language models (LLMs). Our approach uses an LLM to construct a KG from input documents, from which MCQs are then systematically generated. Each MCQ is generated by selecting a node from the KG as the key, sampling a related triple or quintuple—optionally augmented with an extra triple—and prompting an LLM to generate a corresponding stem from these graph components. Distractors are then selected from the KG. For each MCQ, nine difficulty signals are computed and combined into a unified difficulty score using a data-driven approach. Within a 150-MCQ, proof-of-concept dataset from Wikipedia, the proposed signals show interpretable associations with empirical incorrect-answer rates aligning with human responses/performance. The results support the feasibility of the proposed pipeline, yet a larger-scale human study may be required to establish deployment-scale validity. Our approach improves automated MCQ generation by integrating structured knowledge representations with LLMs and a data-driven difficulty estimation model.
- Open Access
- 15 December 2025
- Published by MDPI AG in Journal of Risk and Financial Management
- Vol. 18 (12) , 716
Predicting the Risk of Death for Cryptocurrencies Using Deep Learning
The rapid rise in the popularity of cryptocurrencies has drawn increasing attention from investors, entrepreneurs, and the public in recent years. However, this rapid growth comes with risk: many coins fail early and become what are known as “dead coins”, defined by a lack of recorded activity for more than a year. This study applies deep learning techniques to estimate the short-term risk of a cryptocurrency’s death. Specifically, three Recurrent Neural Network architectures, Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), and Gated Recurrent Unit (GRU), were trained on 18-month time series of daily closing prices and trading volumes using a stratified five-fold cross-validation framework. The models’ predictive performances were compared across input windows ranging from 10 to 180 days. Using the previous 180 days of data as input, GRU achieved the highest point accuracy of 0.7134, whereas BiLSTM exhibited the best performance when evaluated across input sequence lengths varying from 10 to 180 days, reaching an average accuracy of 0.676. These findings show the ability of recurrent architectures to anticipate short-term failure risks in cryptocurrency markets. Theoretically, the study contributes to financial risk modeling by extending time series classification methods to cryptocurrency failure prediction. Practically, it provides investors and analysts with a data-driven early-warning tool to manage portfolio risk and reduce potential losses.
- 15 December 2025
- Published by MDPI AG in Journal of Risk and Financial Management
- Vol. 18 (12) , 716
Open AccessPredicting the Risk of Death for Cryptocurrencies Using Deep Learning
The rapid rise in the popularity of cryptocurrencies has drawn increasing attention from investors, entrepreneurs, and the public in recent years. However, this rapid growth comes with risk: many coins fail early and become what are known as “dead coins”, defined by a lack of recorded activity for more than a year. This study applies deep learning techniques to estimate the short-term risk of a cryptocurrency’s death. Specifically, three Recurrent Neural Network architectures, Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), and Gated Recurrent Unit (GRU), were trained on 18-month time series of daily closing prices and trading volumes using a stratified five-fold cross-validation framework. The models’ predictive performances were compared across input windows ranging from 10 to 180 days. Using the previous 180 days of data as input, GRU achieved the highest point accuracy of 0.7134, whereas BiLSTM exhibited the best performance when evaluated across input sequence lengths varying from 10 to 180 days, reaching an average accuracy of 0.676. These findings show the ability of recurrent architectures to anticipate short-term failure risks in cryptocurrency markets. Theoretically, the study contributes to financial risk modeling by extending time series classification methods to cryptocurrency failure prediction. Practically, it provides investors and analysts with a data-driven early-warning tool to manage portfolio risk and reduce potential losses.
- 12 April 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 318
Multi-label multi-modal classification of movie scenes
- 12 April 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 318
Multi-label multi-modal classification of movie scenes
- Open Access
- 1 September 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Investigating the Validity of Ground Truth in Code Reviewer Recommendation Studies
Background: Selecting the ideal code reviewer in modern code review is a crucial first step to perform effective code reviews. There are several algorithms proposed in the literature for recommending the ideal code reviewer for a given pull request. The success of these code reviewer recommendation algorithms is measured by comparing the recommended reviewers with the ground truth that is the assigned reviewers selected in real life. However, in practice, the assigned reviewer may not be the ideal reviewer for a given pull request.Aims: In this study, we investigate the validity of ground truth data in code reviewer recommendation studies.Method: By conducting an informal literature review, we compared the reviewer selection heuristics in real life and the algorithms used in recommendation models. We further support our claims by using empirical data from code reviewer recommendation studies.Results: By literature review, and accompanying empirical data, we show that ground truth data used in code reviewer recommendation studies is potentially problematic. This reduces the validity of the code reviewer datasets and the reviewer recommendation studies. Conclusion: We demonstrated the cases where the ground truth in code reviewer recommendation studies are invalid and discussed the potential solutions to address this issue.
- 1 September 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Open AccessInvestigating the Validity of Ground Truth in Code Reviewer Recommendation Studies
Background: Selecting the ideal code reviewer in modern code review is a crucial first step to perform effective code reviews. There are several algorithms proposed in the literature for recommending the ideal code reviewer for a given pull request. The success of these code reviewer recommendation algorithms is measured by comparing the recommended reviewers with the ground truth that is the assigned reviewers selected in real life. However, in practice, the assigned reviewer may not be the ideal reviewer for a given pull request.Aims: In this study, we investigate the validity of ground truth data in code reviewer recommendation studies.Method: By conducting an informal literature review, we compared the reviewer selection heuristics in real life and the algorithms used in recommendation models. We further support our claims by using empirical data from code reviewer recommendation studies.Results: By literature review, and accompanying empirical data, we show that ground truth data used in code reviewer recommendation studies is potentially problematic. This reduces the validity of the code reviewer datasets and the reviewer recommendation studies. Conclusion: We demonstrated the cases where the ground truth in code reviewer recommendation studies are invalid and discussed the potential solutions to address this issue.
- 1 October 2017
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 89, 466-486
A window-based time series feature extraction method
- 1 October 2017
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 89, 466-486
A window-based time series feature extraction method
- 1 December 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 375-380
A Privacy-Preserving Solution for the Bipartite Ranking Problem
In this paper, we propose an efficient solution for the privacy-preserving of a bipartite ranking algorithm. The bipartite ranking problem can be considered as finding a function that ranks positive instances (in a dataset) higher than the negative ones. However, one common concern for all the existing schemes is the privacy of individuals in the dataset. That is, one (e.g., a researcher) needs to access the records of all individuals in the dataset in order to run the algorithm. This privacy concern puts limitations on the use of sensitive personal data for such analysis. The RIMARC (Ranking Instances by Maximizing Area under the ROC Curve) algorithm solves the bipartite ranking problem by learning a model to rank instances. As part of the model, it learns weights for each feature by analyzing the area under receiver operating characteristic (ROC) curve. RIMARC algorithm is shown to be more accurate and efficient than its counterparts. Thus, we use this algorithm as a building-block and provide a privacy-preserving version of the RIMARC algorithm using homomorphic encryption and secure multi-party computation. Our proposed algorithm lets a data owner outsource the storage and processing of its encrypted dataset to a semi-trusted cloud. Then, a researcher can get the results of his/her queries (to learn the ranking function) on the dataset by interacting with the cloud. During this process, neither the researcher nor the cloud learns any information about the raw dataset. We prove the security of the proposed algorithm and show its efficiency via experiments on real data.
- 1 December 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 375-380
A Privacy-Preserving Solution for the Bipartite Ranking Problem
In this paper, we propose an efficient solution for the privacy-preserving of a bipartite ranking algorithm. The bipartite ranking problem can be considered as finding a function that ranks positive instances (in a dataset) higher than the negative ones. However, one common concern for all the existing schemes is the privacy of individuals in the dataset. That is, one (e.g., a researcher) needs to access the records of all individuals in the dataset in order to run the algorithm. This privacy concern puts limitations on the use of sensitive personal data for such analysis. The RIMARC (Ranking Instances by Maximizing Area under the ROC Curve) algorithm solves the bipartite ranking problem by learning a model to rank instances. As part of the model, it learns weights for each feature by analyzing the area under receiver operating characteristic (ROC) curve. RIMARC algorithm is shown to be more accurate and efficient than its counterparts. Thus, we use this algorithm as a building-block and provide a privacy-preserving version of the RIMARC algorithm using homomorphic encryption and secure multi-party computation. Our proposed algorithm lets a data owner outsource the storage and processing of its encrypted dataset to a semi-trusted cloud. Then, a researcher can get the results of his/her queries (to learn the ranking function) on the dataset by interacting with the cloud. During this process, neither the researcher nor the cloud learns any information about the raw dataset. We prove the security of the proposed algorithm and show its efficiency via experiments on real data.
- Open Access
- 11 October 2016
- Published by Oxford University Press (OUP) in EP Europace
- Vol. 19 (5) , 741-746
Data mining experiments on the Angiotensin II-Antagonist in Paroxysmal Atrial Fibrillation (ANTIPAF-AFNET 2) trial: ‘exposing the invisible’
The aims of this study include (i) pursuing data-mining experiments on the Angiotensin II-Antagonist in Paroxysmal Atrial Fibrillation (ANTIPAF-AFNET 2) trial dataset containing atrial fibrillation (AF) burden scores of patients with many clinical parameters and (ii) revealing possible correlations between the estimated risk factors of AF and other clinical findings or measurements provided in the dataset. Ranking Instances by Maximizing the Area under a Receiver Operating Characteristics (ROC) Curve (RIMARC) is used to determine the predictive weights (Pw) of baseline variables on the primary endpoint. Chi-square automatic interaction detector algorithm is performed for comparing the results of RIMARC. The primary endpoint of the ANTIPAF-AFNET 2 trial was the percentage of days with documented episodes of paroxysmal AF or with suspected persistent AF. By means of the RIMARC analysis algorithm, baseline SF-12 mental component score (Pw= 0.3597), age (Pw= 0.2865), blood urea nitrogen (BUN) (Pw= 0.2719), systolic blood pressure (Pw= 0.2240), and creatinine level (Pw= 0.1570) of the patients were found to be predictors of AF burden. Atrial fibrillation burden increases as baseline SF-12 mental component score gets lower; systolic blood pressure, BUN and creatinine levels become higher; and the patient gets older. The AF burden increased significantly at age >76. With the ANTIPAF-AFNET 2 dataset, the present data-mining analyses suggest that a baseline SF-12 mental component score, age, systolic blood pressure, BUN, and creatinine level of the patients are predictors of AF burden. Additional studies are necessary to understand the distinct kidney-specific pathophysiological pathways that contribute to AF burden.
- 11 October 2016
- Published by Oxford University Press (OUP) in EP Europace
- Vol. 19 (5) , 741-746
Open AccessData mining experiments on the Angiotensin II-Antagonist in Paroxysmal Atrial Fibrillation (ANTIPAF-AFNET 2) trial: ‘exposing the invisible’
The aims of this study include (i) pursuing data-mining experiments on the Angiotensin II-Antagonist in Paroxysmal Atrial Fibrillation (ANTIPAF-AFNET 2) trial dataset containing atrial fibrillation (AF) burden scores of patients with many clinical parameters and (ii) revealing possible correlations between the estimated risk factors of AF and other clinical findings or measurements provided in the dataset. Ranking Instances by Maximizing the Area under a Receiver Operating Characteristics (ROC) Curve (RIMARC) is used to determine the predictive weights (Pw) of baseline variables on the primary endpoint. Chi-square automatic interaction detector algorithm is performed for comparing the results of RIMARC. The primary endpoint of the ANTIPAF-AFNET 2 trial was the percentage of days with documented episodes of paroxysmal AF or with suspected persistent AF. By means of the RIMARC analysis algorithm, baseline SF-12 mental component score (Pw= 0.3597), age (Pw= 0.2865), blood urea nitrogen (BUN) (Pw= 0.2719), systolic blood pressure (Pw= 0.2240), and creatinine level (Pw= 0.1570) of the patients were found to be predictors of AF burden. Atrial fibrillation burden increases as baseline SF-12 mental component score gets lower; systolic blood pressure, BUN and creatinine levels become higher; and the patient gets older. The AF burden increased significantly at age >76. With the ANTIPAF-AFNET 2 dataset, the present data-mining analyses suggest that a baseline SF-12 mental component score, age, systolic blood pressure, BUN, and creatinine level of the patients are predictors of AF burden. Additional studies are necessary to understand the distinct kidney-specific pathophysiological pathways that contribute to AF burden.
- Open Access
- 2 July 2016
- Published by Oxford University Press (OUP) in EP Europace
- Vol. 19 (6) , 921-928
Predictors of sinus rhythm after electrical cardioversion of atrial fibrillation: results from a data mining project on the Flec-SL trial data set
Data mining is the computational process to obtain information from a data set and transform it for further use. Herein, through data mining with supportive statistical analyses, we identified and consolidated variables of the Flecainide Short-Long (Flec-SL—AFNET 3) trial dataset that are associated with the primary outcome of the trial, recurrence of persistent atrial fibrillation (AF) or death. The ‘Ranking Instances by Maximizing the Area under the ROC Curve’ (RIMARC) algorithm was applied to build a classifier that can predict the primary outcome by using variables in the Flec-SL dataset. The primary outcome was time to persistent AF or death. The RIMARC algorithm calculated the predictive weights of each variable in the Flec-SL dataset for the primary outcome. Among the initial 21 parameters, 6 variables were identified by the RIMARC algorithm. In univariate Cox regression analysis of these variables, increased heart rate during AF and successful pharmacological conversion (PC) to sinus rhythm (SR) were found to be significant predictors. Multivariate Cox regression analysis revealed successful PC as the single relevant predictor of SR maintenance. The primary outcome risk was 3.14 times (95% CI:1.7–5.81) lower in those who had successful PC to SR than those who needed electrical cardioversion. Pharmacological conversion of persistent AF with flecainide without the need for electrical cardioversion is a powerful and independent predictor of maintenance of SR. A strategy of flecainide pretreatment for 48 h prior to planned electrical cardioversion may be a useful planning of a strategy of long-term rhythm control.
- 2 July 2016
- Published by Oxford University Press (OUP) in EP Europace
- Vol. 19 (6) , 921-928
Open AccessPredictors of sinus rhythm after electrical cardioversion of atrial fibrillation: results from a data mining project on the Flec-SL trial data set
Data mining is the computational process to obtain information from a data set and transform it for further use. Herein, through data mining with supportive statistical analyses, we identified and consolidated variables of the Flecainide Short-Long (Flec-SL—AFNET 3) trial dataset that are associated with the primary outcome of the trial, recurrence of persistent atrial fibrillation (AF) or death. The ‘Ranking Instances by Maximizing the Area under the ROC Curve’ (RIMARC) algorithm was applied to build a classifier that can predict the primary outcome by using variables in the Flec-SL dataset. The primary outcome was time to persistent AF or death. The RIMARC algorithm calculated the predictive weights of each variable in the Flec-SL dataset for the primary outcome. Among the initial 21 parameters, 6 variables were identified by the RIMARC algorithm. In univariate Cox regression analysis of these variables, increased heart rate during AF and successful pharmacological conversion (PC) to sinus rhythm (SR) were found to be significant predictors. Multivariate Cox regression analysis revealed successful PC as the single relevant predictor of SR maintenance. The primary outcome risk was 3.14 times (95% CI:1.7–5.81) lower in those who had successful PC to SR than those who needed electrical cardioversion. Pharmacological conversion of persistent AF with flecainide without the need for electrical cardioversion is a powerful and independent predictor of maintenance of SR. A strategy of flecainide pretreatment for 48 h prior to planned electrical cardioversion may be a useful planning of a strategy of long-term rhythm control.
- Open Access
- 10 December 2015
- Published by Taylor & Francis in Gynecological Endocrinology
- Vol. 32 (5) , 361-365
Progesterone change in the late follicular phase affects pregnancy rates both agonist and antagonist protocols in normoresponders: a case-controlled study in ICSI cycles
Objective: The aim of the presented study is to investigate the impact of progesterone change in the late follicular phase on the pregnancy rates of both agonist and antagonist protocols in normoresponders. Study design: A total of 201 normoresponder patients, who underwent embryo transfer were consecutively selected. 118 patients were stimulated using a long luteal GnRH agonist protocol and 83 using a flexible antagonist protocol. The level of change in late follicular phase progesterone was calculated according to the progesterone levels on the hCG day and pre-hCG day (1 or 2 days prior to hCG day) measurement. Results: Clinical pregnancy rates were comparable between long luteal and antagonist group (35.6 and 41%, respectively). The incidence of progesterone elevation on the hCG day was 11% in long luteal and 18% in antagonist group (p = 0.16). In pregnant cycles, p levels both on the hCG day and pre-hCG day measurement were significantly higher in antagonist than agonist cycles (p = 0.029, p = 0.038, respectively). The change of p level was statistically significant in non-pregnant cycles both for the agonist (-0.17 ± 0.07; 95% CI: −0.29 to −0.37) and antagonist groups (−0.18 ± 0.07; 95%CI: −0.31 to −0.04). Conclusions: Late follicular phase progesterone levels were stable during the cycles of pregnant patients irrespective of the protocols and were shown to be higher in pregnant patients in antagonist cycles when compared to agonist cycles.
- 10 December 2015
- Published by Taylor & Francis in Gynecological Endocrinology
- Vol. 32 (5) , 361-365
Open AccessProgesterone change in the late follicular phase affects pregnancy rates both agonist and antagonist protocols in normoresponders: a case-controlled study in ICSI cycles
Objective: The aim of the presented study is to investigate the impact of progesterone change in the late follicular phase on the pregnancy rates of both agonist and antagonist protocols in normoresponders. Study design: A total of 201 normoresponder patients, who underwent embryo transfer were consecutively selected. 118 patients were stimulated using a long luteal GnRH agonist protocol and 83 using a flexible antagonist protocol. The level of change in late follicular phase progesterone was calculated according to the progesterone levels on the hCG day and pre-hCG day (1 or 2 days prior to hCG day) measurement. Results: Clinical pregnancy rates were comparable between long luteal and antagonist group (35.6 and 41%, respectively). The incidence of progesterone elevation on the hCG day was 11% in long luteal and 18% in antagonist group (p = 0.16). In pregnant cycles, p levels both on the hCG day and pre-hCG day measurement were significantly higher in antagonist than agonist cycles (p = 0.029, p = 0.038, respectively). The change of p level was statistically significant in non-pregnant cycles both for the agonist (-0.17 ± 0.07; 95% CI: −0.29 to −0.37) and antagonist groups (−0.18 ± 0.07; 95%CI: −0.31 to −0.04). Conclusions: Late follicular phase progesterone levels were stable during the cycles of pregnant patients irrespective of the protocols and were shown to be higher in pregnant patients in antagonist cycles when compared to agonist cycles.
- Open Access
- 17 April 2015
- Published by Springer Nature in Medical & Biological Engineering & Computing
- Vol. 53 (9) , 911-920
Estimating the chance of success in IVF treatment using a ranking algorithm
In medicine, estimating the chance of success for treatment is important in deciding whether to begin the treatment or not. This paper focuses on the domain of in vitro fertilization (IVF), where estimating the outcome of a treatment is very crucial in the decision to proceed with treatment for both the clinicians and the infertile couples. IVF treatment is a stressful and costly process. It is very stressful for couples who want to have a baby. If an initial evaluation indicates a low pregnancy rate, decision of the couple may change not to start the IVF treatment. The aim of this study is twofold, firstly, to develop a technique that can be used to estimate the chance of success for a couple who wants to have a baby and secondly, to determine the attributes and their particular values affecting the outcome in IVF treatment. We propose a new technique, called success estimation using a ranking algorithm (SERA), for estimating the success of a treatment using a ranking-based algorithm. The particular ranking algorithm used here is RIMARC. The performance of the new algorithm is compared with two well-known algorithms that assign class probabilities to query instances. The algorithms used in the comparison are Naïve Bayes Classifier and Random Forest. The comparison is done in terms of area under the ROC curve, accuracy and execution time, using tenfold stratified cross-validation. The results indicate that the proposed SERA algorithm has a potential to be used successfully to estimate the probability of success in medical treatment.
- 17 April 2015
- Published by Springer Nature in Medical & Biological Engineering & Computing
- Vol. 53 (9) , 911-920
Open AccessEstimating the chance of success in IVF treatment using a ranking algorithm
In medicine, estimating the chance of success for treatment is important in deciding whether to begin the treatment or not. This paper focuses on the domain of in vitro fertilization (IVF), where estimating the outcome of a treatment is very crucial in the decision to proceed with treatment for both the clinicians and the infertile couples. IVF treatment is a stressful and costly process. It is very stressful for couples who want to have a baby. If an initial evaluation indicates a low pregnancy rate, decision of the couple may change not to start the IVF treatment. The aim of this study is twofold, firstly, to develop a technique that can be used to estimate the chance of success for a couple who wants to have a baby and secondly, to determine the attributes and their particular values affecting the outcome in IVF treatment. We propose a new technique, called success estimation using a ranking algorithm (SERA), for estimating the success of a treatment using a ranking-based algorithm. The particular ranking algorithm used here is RIMARC. The performance of the new algorithm is compared with two well-known algorithms that assign class probabilities to query instances. The algorithms used in the comparison are Naïve Bayes Classifier and Random Forest. The comparison is done in terms of area under the ROC curve, accuracy and execution time, using tenfold stratified cross-validation. The results indicate that the proposed SERA algorithm has a potential to be used successfully to estimate the probability of success in medical treatment.