scholarly journals Not all biases are bad: equitable and inequitable biases in machine learning and radiology

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Mirjam Pot ◽  
Nathalie Kieusseyan ◽  
Barbara Prainsack
Keyword(s):  
Machine Learning ◽  
Human Error ◽  
Technical Problem ◽  
Social Dimension ◽  
Patient Treatment ◽  
Technical Solution ◽  
Negative Effects ◽  
Different Types ◽  
Human Decision

AbstractThe application of machine learning (ML) technologies in medicine generally but also in radiology more specifically is hoped to improve clinical processes and the provision of healthcare. A central motivation in this regard is to advance patient treatment by reducing human error and increasing the accuracy of prognosis, diagnosis and therapy decisions. There is, however, also increasing awareness about bias in ML technologies and its potentially harmful consequences. Biases refer to systematic distortions of datasets, algorithms, or human decision making. These systematic distortions are understood to have negative effects on the quality of an outcome in terms of accuracy, fairness, or transparency. But biases are not only a technical problem that requires a technical solution. Because they often also have a social dimension, the ‘distorted’ outcomes they yield often have implications for equity. This paper assesses different types of biases that can emerge within applications of ML in radiology, and discusses in what cases such biases are problematic. Drawing upon theories of equity in healthcare, we argue that while some biases are harmful and should be acted upon, others might be unproblematic and even desirable—exactly because they can contribute to overcome inequities.

scholarly journals Machine Learning Approach to Classify Rain Type Based on Thies Disdrometers and Cloud Observations

Atmosphere ◽  
2019 ◽  
Vol 10 (5) ◽  
pp. 251 ◽  
Author(s):  
Wael Ghada ◽  
Nicole Estrella ◽  
Annette Menzel
Keyword(s):  
Machine Learning ◽  
Geographical Location ◽  
Classification Performance ◽  
Classification Methods ◽  
Reference Dataset ◽  
Machine Learning Classification ◽  
Machine Learning Approach ◽  
Different Types ◽  
Microstructure Parameters

Rain microstructure parameters assessed by disdrometers are commonly used to classify rain into convective and stratiform. However, different types of disdrometer result in different values for these parameters. This in turn potentially deteriorates the quality of rain type classifications. Thies disdrometer measurements at two sites in Bavaria in southern Germany were combined with cloud observations to construct a set of clear convective and stratiform intervals. This reference dataset was used to study the performance of classification methods from the literature based on the rain microstructure. We also explored the possibility of improving the performance of these methods by tuning the decision boundary. We further identified highly discriminant rain microstructure parameters and used these parameters in five machine-learning classification models. Our results confirm the potential of achieving high classification performance by applying the concepts of machine learning compared to already available methods. Machine-learning classification methods provide a concrete and flexible procedure that is applicable regardless of the geographical location or the device. The suggested procedure for classifying rain types is recommended prior to studying rain microstructure variability or any attempts at improving radar estimations of rain intensity.

scholarly journals An IHE-conform telecooperation platform supporting the treatment of dementia patients

2015 ◽  
Vol 1 (1) ◽  
pp. 310-313
Author(s):  
K. Saleh ◽  
D. Ammon ◽  
S. Lehnert ◽  
S. Röhr ◽  
V. Detschew ◽  
...  
Keyword(s):  
Health Care Professionals ◽  
Rural Areas ◽  
Patient Treatment ◽  
Electronic Patient Records ◽  
Medical Needs ◽  
Intersectoral Cooperation ◽  
Dementia Patients ◽  
Different Types ◽  
Unmet Medical Needs

AbstractEnsuring medical support of patients of advanced age in rural areas is a major challenge. Moreover, the number of registered doctors—medical specialists in particular—will decrease in such areas over the next years. These unmet medical needs in combination with communication deficiencies among different types of health-care professionals pose threats to the quality of patient treatment. This work presents a novel solution combining telemedicine, telecooperation, and IHE profiles to tackle these challenges. We present a telecooperation platform that supports longitudinal electronic patient records and allows for intersectoral cooperation based on shared electronic medication charts and other documents. Furthermore, the conceived platform allows for an integration into the planned German telematics infrastructure.

scholarly journals Abduction-Based Explanations for Machine Learning Models

2019 ◽  
Vol 33 ◽  
pp. 1511-1519 ◽  
Author(s):  
Alexey Ignatiev ◽  
Nina Narodytska ◽  
Joao Marques-Silva
Keyword(s):  
Machine Learning ◽  
Decision Problem ◽  
Abductive Reasoning ◽  
Decision Makers ◽  
Reasoning System ◽  
Constraint Reasoning ◽  
Human Decision ◽  
Applications Of Machine Learning ◽  
Machine Learning Models

The growing range of applications of Machine Learning (ML) in a multitude of settings motivates the ability of computing small explanations for predictions made. Small explanations are generally accepted as easier for human decision makers to understand. Most earlier work on computing explanations is based on heuristic approaches, providing no guarantees of quality, in terms of how close such solutions are from cardinality- or subset-minimal explanations. This paper develops a constraint-agnostic solution for computing explanations for any ML model. The proposed solution exploits abductive reasoning, and imposes the requirement that the ML model can be represented as sets of constraints using some target constraint reasoning system for which the decision problem can be answered with some oracle. The experimental results, obtained on well-known datasets, validate the scalability of the proposed approach as well as the quality of the computed solutions.

scholarly journals On the influence of template size, canonicalization and exclusivity for retrosynthesis and reaction prediction applications

2021 ◽  
Author(s):  
Esther Heid ◽  
Jiannan Liu ◽  
Andrea Aude ◽  
William H. Green
Keyword(s):  
Machine Learning ◽  
Computationally Efficient ◽  
Negative Effects ◽  
Reaction Prediction ◽  
Ranking Algorithms ◽  
Ranking Models ◽  
Data Driven Approach ◽  
Important Basis ◽  
Overlapping Sets

Heuristic and machine learning models for rank-ordering reaction templates comprise an important basis for computer-aided organic synthesis regarding both product prediction and retrosynthetic pathway planning. Their viability relies heavily on the quality and characteristics of the underlying template database. With the advent of automated reaction and template extraction software and consequently the creation of template databases too large to be curated manually, a data-driven approach to assess and improve the quality of template sets is needed. We therefore systematically studied the influence of template generality, canonicalization and exclusivity on the performance of different template ranking models. We find that duplicate and non-exclusive templates, \textit{i.e.} templates which describe the same chemical transformation on identical or overlapping sets of molecules, decrease both the accuracy of the ranking algorithm and the applicability of the respective top-ranked templates significantly. To remedy the negative effects of non-exclusivity, we developed a general and computationally efficient framework to deduplicate and hierarchically correct templates. As a result, performance improved for both heuristic and machine learning template ranking algorithms across different template sizes. The canonicalization and correction code was made freely available.

scholarly journals Animal Fats in Rabbit Feeding – A Review

2020 ◽  
Vol 20 (4) ◽  
pp. 1185-1215
Author(s):  
Andrzej Gugołek ◽  
Dorota Kowalska
Keyword(s):  
Growth Performance ◽  
Scientific Literature ◽  
Insect Larvae ◽  
Negative Effects ◽  
Animal Fats ◽  
Pork Lard ◽  
Poultry Fat ◽  
Different Types ◽  
History Of

AbstractThe purpose of this article is to overview the history of feeding rabbits with different types of animal fats, and to discuss their effects on rabbit performance and quality of their products. Other aspects of the inclusion of various animal fats in rabbit diets are also described. This article is based on the analysis of relevant scientific literature and presents animal fats fed to rabbits, such as beef tallow, butter, pork lard, poultry fat, fish oil, krill oil, oil extracted from insect larvae, mixtures of various animal fats, and mixtures of animal and vegetable fats. The reported papers describe the effect of fats on growth performance, lactation, rearing performance, meat quality, and health status of rabbits. It is notable that in many cases, various animal fats were often an integral part of numerous diets or were included in control diets. The presented information demonstrates that animal fat can be fed to rabbits at 2–4% of the diet without negative effects on reproductive performance, growth performance and quality of meat obtained. Rabbits were used as model animals in many studies in which fat was added to balance the diets and to increase their energy value, especially when investigating various cardiovascular and obesity-related diseases.

Can trust become a factor of economic growth? Dynamic changes in the level of trust of Russian youth

2020 ◽  
pp. 92-107 ◽  
Author(s):  
A. I. Bakhtigaraeva ◽  
A. A. Stavinskaya
Keyword(s):  
Institutional Environment ◽  
Negative Effects ◽  
Dynamic Changes ◽  
Generalized Trust ◽  
Level Of Education ◽  
Advantages And Disadvantages ◽  
The Future ◽  
Analysis Of Dynamics

The article considers the role of trust in the economy, the mechanisms of its accumulation and the possibility of using it as one of the growth factors in the future. The advantages and disadvantages of measuring the level of generalized trust using two alternative questions — about trusting people in general and trusting strangers — are analyzed. The results of the analysis of dynamics of the level of generalized trust among Russian youth, obtained within the study of the Institute for National Projects in 10 regions of Russia, are presented. It is shown that there are no significant changes in trust in people in general during the study at university. At the same time, the level of trust in strangers falls, which can negatively affect the level of trust in the country as a whole, and as a result have negative effects on the development of the economy in the future. Possible causes of the observed trends and the role of universities are discussed. Also the question about the connection between the level of education and generalized trust in countries with different quality of the institutional environment is raised.

A Literature Review Study of Software Defect Prediction using Machine Learning Techniques

2018 ◽  
Vol 6 (6) ◽  
pp. 300 ◽  
Author(s):  
Feidu Akmel ◽  
Ermiyas Birihanu ◽  
Bahir Siraj
Keyword(s):  
Machine Learning ◽  
Software Metrics ◽  
Quality Standard ◽  
Machine Learning Techniques ◽  
Software Systems ◽  
Health Care Insurance ◽  
Software Defect ◽  
Learning Techniques ◽  
Software Product

Software systems are any software product or applications that support business domains such as Manufacturing,Aviation, Health care, insurance and so on.Software quality is a means of measuring how software is designed and how well the software conforms to that design. Some of the variables that we are looking for software quality are Correctness, Product quality, Scalability, Completeness and Absence of bugs, However the quality standard that was used from one organization is different from other for this reason it is better to apply the software metrics to measure the quality of software. Attributes that we gathered from source code through software metrics can be an input for software defect predictor. Software defect are an error that are introduced by software developer and stakeholders. Finally, in this study we discovered the application of machine learning on software defect that we gathered from the previous research works.

Sign in / Sign up

Export Citation Format

Share Document