Discovering diagnostic rules from a neurotologic database with genetic algorithms

Machine Learning in the Management of Lateral Skull Base Tumors: A Systematic Review

The application of machine learning (ML) techniques to otolaryngology remains a topic of interest and prevalence in the literature, though no previous articles have summarized the current state of ML application to management and the diagnosis of lateral skull base (LSB) tumors. Subsequently, we present a systematic overview of previous applications of ML techniques to the management of LSB tumors. Independent searches were conducted on PubMed and Web of Science between August 2020 and February 2021 to identify the literature pertaining to the use of ML techniques in LSB tumor surgery written in the English language. All articles were assessed in regard to their application task, ML methodology, and their outcomes. A total of 32 articles were examined. The number of articles involving applications of ML techniques to LSB tumor surgeries has significantly increased since the first article relevant to this field was published in 1994. The most commonly employed ML category was tree-based algorithms. Most articles were included in the category of surgical management (13; 40.6%), followed by those in disease classification (8; 25%). Overall, the application of ML techniques to the management of LSB tumor has evolved rapidly over the past two decades, and the anticipated growth in the future could significantly augment the surgical outcomes and management of LSB tumors.

Machine Learning Techniques for Differential Diagnosis of Vertigo and Dizziness: A Review

Vertigo is a sensation of movement that results from disorders of the inner ear balance organs and their central connections, with aetiologies that are often benign and sometimes serious. An individual who develops vertigo can be effectively treated only after a correct diagnosis of the underlying vestibular disorder is reached. Recent advances in artificial intelligence promise novel strategies for the diagnosis and treatment of patients with this common symptom. Human analysts may experience difficulties manually extracting patterns from large clinical datasets. Machine learning techniques can be used to visualize, understand, and classify clinical data to create a computerized, faster, and more accurate evaluation of vertiginous disorders. Practitioners can also use them as a teaching tool to gain knowledge and valuable insights from medical data. This paper provides a review of the literatures from 1999 to 2021 using various feature extraction and machine learning techniques to diagnose vertigo disorders. This paper aims to provide a better understanding of the work done thus far and to provide future directions for research into the use of machine learning in vertigo diagnosis.

A Systematic Review of Clinical Vestibular Symptom Triage, Tools, and Algorithms

OBJECTIVE

The evaluation of peripheral vestibular disorders in clinical practice is an especially difficult endeavor, particularly for the inexperienced clinician. The goal of this systematic review is thus to evaluate the design, approaches, and outcomes for clinical vestibular symptom triage and decision support tools reported in contemporary published literature.

DATA SOURCES

A comprehensive search of existing literature in August 2020 was conducted using MEDLINE, CINAHL, and EMBASE using terms of desired diagnostic tools such as algorithm, protocol, and questionnaire as well as an exhaustive set of terms to encompass vestibular disorders.

REVIEW METHODS

Study characteristics, tool metrics, and performance were extracted using a standardized form. Quality assessment was conducted using a modified version of the Quality of Diagnostic Accuracy Studies 2 (QUADAS-2) assessment tool.

RESULTS

A total of 18 articles each reporting a novel tool for the evaluation of vestibular disorders were identified. Tools were organized into 3 discrete categories, including self-administered questionnaires, health care professional administered tools, and decision support systems. Most tools could differentiate between specific vestibular pathologies, with outcome measures including sensitivity, specificity, and accuracy.

CONCLUSION

A multitude of tools have been published to aid with the evaluation of vertiginous patients. Our systematic review identified several low-evidence reports of triage and decision support tools for the evaluation of vestibular disorders.

Clinical Practice Guideline: Benign Paroxysmal Positional Vertigo (Update)

Objective

This update of a 2008 guideline from the American Academy of Otolaryngology—Head and Neck Surgery Foundation provides evidence-based recommendations to benign paroxysmal positional vertigo (BPPV), defined as a disorder of the inner ear characterized by repeated episodes of positional vertigo. Changes from the prior guideline include a consumer advocate added to the update group; new evidence from 2 clinical practice guidelines, 20 systematic reviews, and 27 randomized controlled trials; enhanced emphasis on patient education and shared decision making; a new algorithm to clarify action statement relationships; and new and expanded recommendations for the diagnosis and management of BPPV.

Purpose

The primary purposes of this guideline are to improve the quality of care and outcomes for BPPV by improving the accurate and efficient diagnosis of BPPV, reducing the inappropriate use of vestibular suppressant medications, decreasing the inappropriate use of ancillary testing such as radiographic imaging, and increasing the use of appropriate therapeutic repositioning maneuvers. The guideline is intended for all clinicians who are likely to diagnose and manage patients with BPPV, and it applies to any setting in which BPPV would be identified, monitored, or managed. The target patient for the guideline is aged ≥18 years with a suspected or potential diagnosis of BPPV. The primary outcome considered in this guideline is the resolution of the symptoms associated with BPPV. Secondary outcomes considered include an increased rate of accurate diagnoses of BPPV, a more efficient return to regular activities and work, decreased use of inappropriate medications and unnecessary diagnostic tests, reduction in recurrence of BPPV, and reduction in adverse events associated with undiagnosed or untreated BPPV. Other outcomes considered include minimizing costs in the diagnosis and treatment of BPPV, minimizing potentially unnecessary return physician visits, and maximizing the health-related quality of life of individuals afflicted with BPPV.

Action Statements

The update group made strong recommendations that clinicians should (1) diagnose posterior semicircular canal BPPV when vertigo associated with torsional, upbeating nystagmus is provoked by the Dix-Hallpike maneuver, performed by bringing the patient from an upright to supine position with the head turned 45° to one side and neck extended 20° with the affected ear down, and (2) treat, or refer to a clinician who can treat, patients with posterior canal BPPV with a canalith repositioning procedure. The update group made a strong recommendation against postprocedural postural restrictions after canalith repositioning procedure for posterior canal BPPV. The update group made recommendations that the clinician should (1) perform, or refer to a clinician who can perform, a supine roll test to assess for lateral semicircular canal BPPV if the patient has a history compatible with BPPV and the Dix-Hallpike test exhibits horizontal or no nystagmus; (2) differentiate, or refer to a clinician who can differentiate, BPPV from other causes of imbalance, dizziness, and vertigo; (3) assess patients with BPPV for factors that modify management, including impaired mobility or balance, central nervous system disorders, a lack of home support, and/or increased risk for falling; (4) reassess patients within 1 month after an initial period of observation or treatment to document resolution or persistence of symptoms; (5) evaluate, or refer to a clinician who can evaluate, patients with persistent symptoms for unresolved BPPV and/or underlying peripheral vestibular or central nervous system disorders; and (6) educate patients regarding the impact of BPPV on their safety, the potential for disease recurrence, and the importance of follow-up. The update group made recommendations against (1) radiographic imaging for a patient who meets diagnostic criteria for BPPV in the absence of additional signs and/or symptoms inconsistent with BPPV that warrant imaging, (2) vestibular testing for a patient who meets diagnostic criteria for BPPV in the absence of additional vestibular signs and/or symptoms inconsistent with BPPV that warrant testing, and (3) routinely treating BPPV with vestibular suppressant medications such as antihistamines and/or benzodiazepines. The guideline update group provided the options that clinicians may offer (1) observation with follow-up as initial management for patients with BPPV and (2) vestibular rehabilitation, either self-administered or with a clinician, in the treatment of BPPV.

Disease Profiling for Computerized Peer Support of Ménière's Disease

Background

Peer support is an emerging form of person-driven active health care. Chronic conditions such as Ménière’s disease (a disorder of the inner ear) need continuing rehabilitation and support that is beyond the scope of routine clinical medical practice. Hence, peer-support programs can be helpful in supplementing some of the rehabilitation aspects.

Objective

The aim of this study was to design a computerized data collection system for the peer support of Menière’s disease that is capable in profiling the subject for diagnosis and in assisting with problem solving.

Methods

The expert program comprises several data entries focusing on symptoms, activity limitations, participation restrictions, quality of life, attitude and personality trait, and an evaluation of disease-specific impact. Data was collected from 740 members of the Finnish Ménière’s Federation and utilized in the construction and evaluation of the program.

Results

The program verifies the diagnosis of a person by using an expert system, and the inference engine selects 50 cases with matched symptom severity by using a nearest neighbor algorithm. These cases are then used as a reference group to compare with the person’s attitude, sense of coherence, and anxiety. The program provides feedback for the person and uses this information to guide the person through the problem-solving process.

Conclusions

This computer-based peer-support program is the first example of an advanced computer-oriented approach using artificial intelligence, both in the profiling of the disease and in profiling the person’s complaints for hearing loss, tinnitus, and vertigo.

Rapid Systematic Review of Normal Audiometry Results as a Predictor for Benign Paroxysmal Positional Vertigo

Objective

To evaluate whether absence of hearing loss on pure-tone audiometry (PTA) is reliable as a diagnostic test for predicting benign paroxysmal positional vertigo (BPPV) in adult patients with vertigo.

Data Sources

PubMed, Embase, and the Cochrane Library.

Methods

A systematic literature search was conducted on December 10, 2013. Relevant publications were selected based on title, abstract, and full text. Selected articles were assessed for relevance and risk of bias using predetermined criteria. Prevalence and the positive and negative predictive value (PPV and NPV) were extracted.

Results

Of 603 retrieved publications, 1 article with high relevance and moderate risk of bias was included. In this study, the prevalence of BPPV was 28%. The PPV of hearing loss assessed by PTA was 31% (95% CI, 17-49) and the NPV was 73% (95% CI, 61-83). The absence of hearing loss on PTA decreased the risk of BPPV by 1%.

Conclusion and Recommendation

There is insufficient high-quality evidence regarding the diagnostic value of the absence of hearing loss, assessed by PTA, for predicting BPPV in adult patients with vertigo.

Use of ICF in assessing the effects of Meniere's disorder on life

OBJECTIVES

We sought to determine the value of the World Health Organization's International Classification of Functioning, Disability and Health (ICF) in subjects with Meniere's disorder in relation to their quality of life.

METHODS

We asked 228 members of the Finnish Meniere Federation to report the effects that Meniere's disorder had on their lives. The replies were classified on the basis of the ICF classification and related to the EuroQol 5D score and disease-specific impact. Logistic regression and decision tree analyses were used to determine the relationships.

RESULTS

Seventy percent of the patients listed impairments, 39% activity limitations, 47% participation restrictions, 16% effects on environmental contextual factors, and 28% effects on personal contextual factors. The EuroQol 5D score was explained by reported vertigo, anxiety, fatigue, restriction of life, and communication problems. The disease-specific impact was explained by episodes of vertigo, fatigue, communication problems, inability to work, restriction of life, and uncertainty of life. Both analysis models provided the same outcome variables, although the decision tree separated the results better (80%) into correct classes than did logistic regression analysis (60%).

CONCLUSIONS

Self-reported participation restriction, activity limitation, and personal contextual factors describe the limitations of general life in subjects with Meniere's disorder. The use of the ICF classification provides an instrument that can be used in enablement of subjects with Meniere's disorder.

A prediction model based on an artificial intelligence system for moderate to severe obstructive sleep apnea

STUDY OBJECTIVES

Obstructive sleep apnea (OSA) is a major concern in modern medicine; however, it is difficult to diagnose. Screening questionnaires such as the Berlin questionnaire, Rome questionnaire, and BASH'IM score are used to identify patients with OSA. However, the sensitivity and specificity of these tools are not satisfactory. We aim to introduce an artificial intelligence method to screen moderate to severe OSA patients (apnea-hypopnea index ≧15).

PATIENTS AND METHODS

One hundred twenty patients were asked to complete a newly developed questionnaire before undergoing an overnight polysomnography (PSG) study. One hundred ten validated questionnaires were enrolled in this study. Genetic algorithm (GA) was used to build the five best models based on these questionnaires. The same data were analyzed with logistic regression (LR) for comparison.

RESULTS

The sensitivity of the GA models varied from 81.8% to 88.0%, with a specificity of 95% to 97%. On the other hand, the sensitivity and specificity of the LR model were 55.6% and 57.9%, respectively.

CONCLUSIONS

GA provides a good solution to build models for screening moderate to severe OSA patients, who require PSG evaluation and medical intervention. The questionnaire did not require any special biochemistry data and was easily self-administered. The sensitivity and specificity of the GA models are satisfactory and may improve when more patients are recruited.

Clinical Practice Guideline: Benign Paroxysmal Positional Vertigo

Objectives:

This guideline provides evidence-based recommendations on managing benign paroxysmal positional vertigo (BPPV), which is the most common vestibular disorder in adults, with a lifetime prevalence of 2.4 percent. The guideline targets patients aged 18 years or older with a potential diagnosis of BPPV, evaluated in any setting in which an adult with BPPV would be identified, monitored, or managed. This guideline is intended for all clinicians who are likely to diagnose and manage adults with BPPV.

Purpose:

The primary purposes of this guideline are to improve quality of care and outcomes for BPPV by improving the accurate and efficient diagnosis of BPPV, reducing the inappropriate use of vestibular suppressant medications, decreasing the inappropriate use of ancillary tests such as radiographic imaging and vestibular testing, and to promote the use of effective repositioning maneuvers for treatment. In creating this guideline, the American Academy of Otolaryngology—Head and Neck Surgery Foundation selected a panel representing the fields of audiology, chiropractic medicine, emergency medicine, family medicine, geriatric medicine, internal medicine, neurology, nursing, otolaryngology–head and neck surgery, physical therapy, and physical medicine and rehabilitation.

Results

The panel made strong recommendations that 1) clinicians should diagnose posterior semicircular canal BPPV when vertigo associated with nystagmus is provoked by the Dix-Hallpike maneuver. The panel made recommendations against 1) radiographic imaging, vestibular testing, or both in patients diagnosed with BPPV, unless the diagnosis is uncertain or there are additional symptoms or signs unrelated to BPPV that warrant testing; and 2) routinely treating BPPV with vestibular suppressant medications such as antihistamines or benzodiazepines. The panel made recommendations that 1) if the patient has a history compatible with BPPV and the Dix-Hallpike test is negative, clinicians should perform a supine roll test to assess for lateral semicircular canal BPPV; 2) clinicians should differentiate BPPV from other causes of imbalance, dizziness, and vertigo; 3) clinicians should question patients with BPPV for factors that modify management including impaired mobility or balance, CNS disorders, lack of home support, and increased risk for falling; 4) clinicians should treat patients with posterior canal BPPV with a particle repositioning maneuver (PRM); 5) clinicians should reassess patients within 1 month after an initial period of observation or treatment to confirm symptom resolution; 6) clinicians should evaluate patients with BPPV who are initial treatment failures for persistent BPPV or underlying peripheral vestibular or CNS disorders; and 7) clinicians should counsel patients regarding the impact of BPPV on their safety, the potential for disease recurrence, and the importance of follow-up. The panel offered as options that 1) clinicians may offer vestibular rehabilitation, either self-administered or with a clinician, for the initial treatment of BPPV and 2) clinicians may offer observation as initial management for patients with BPPV and with assurance of follow-up. The panel made no recommendation concerning audiometric testing in patients diagnosed with BPPV.

Disclaimer:

This clinical practice guideline is not intended as a sole source of guidance in managing benign paroxysmal positional vertigo. Rather, it is designed to assist clinicians by providing an evidence-based framework for decision-making strategies. The guideline is not intended to replace clinical judgement or establish a protocol for all individuals with this condition, and may not provide the only appropriate approach to diagnosing and managing this problem. ® 2008 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved.

Evolving Structure-Function Mappings in Cognitive Neuroscience Using Genetic Programming

A challenging goal of psychology and neuroscience is to map cognitive functions onto neuroanatomical structures. This paper shows how computational methods based upon evolutionary algorithms can facilitate the search for satisfactory mappings by efficiently combining constraints from neuroanatomy and physiology (the structures) with constraints from behavioural experiments (the functions). This methodology involves creation of a database coding for known neuroanatomical and physiological constraints, for mental programs made of primitive cognitive functions, and for typical experiments with their behavioural results. The evolutionary algorithms evolve theories mapping structures to functions in order to optimize the fit with the actual data. These theories lead to new, empirically testable predictions. The role of the prefrontal cortex in humans is discussed as an example. This methodology can be applied to the study of structures or functions alone, and can also be used to study other complex systems.

Evaluation of capnography using a genetic algorithm to predict PaCO2

INTRODUCTION

Noninvasive estimates of Paco(2) are usually done by measuring exhaled carbon dioxide at end-expiration (Petco(2)). While commonly used in studies involving healthy patients, it is less useful in sicker patients. Conditions that affect the terminal dead space and hence the accuracy of Petco(2) as a surrogate for Paco(2) may also affect other components of the capnogram. A genetic algorithm is a computer technique for discovering relationships between variables. The purpose of this study was to use a genetic algorithm to improve the precision of Paco(2) prediction in comparison to Petco(2).

METHODS

Inspiratory and expiratory volumes were measured and analyzed by the computerized capnogram. Data were recorded for 2 min. Within 5 min of recording the capnograms, arterial blood gases were obtained. After excluding artifact and incomplete capnograms, five of the remaining breaths from each patient were selected. A genetic algorithm, constructed in postfix notation, consisted of 1,000 chromosomes with genes randomly selected from the 11 capnographic data fields and mathematical operators. The algorithm was constructed on 400 breaths from 83 randomly selected patients (construction group) and tested on 160 breaths from the remaining 32 patients (test group).

RESULTS

For the construction group, the bias and precision between Petco(2) and Paco(2) were 4.3 +/- 4.9 mm Hg (mean +/- SD). For the 160 breaths in the test group, Petco(2) predicted Paco(2) with bias and precision of 2.9 +/- 4.2 mm Hg. The best chromosome found by the genetic algorithm was (10 x 5 + 5 x 5 x 5)/(10 x 10) x Petco(2) - (5 x 5 x 10 + 5 x 5)/(10 x 10) x int time + 2 x 2 x 2 x 2 + (2 x 2)/10, which reduces to 0.65 x Petco(2) - 2.75 x int time + 16.4. This produced a bias and precision of 0.9 +/- 4.1 mm Hg in the construction group and 0 +/- 3.7 mm Hg in the test group (p < 0.01).

CONCLUSIONS

In this study of nonintubated emergency department patients, a genetic algorithm produced an improvement in bias and precision of Paco(2) prediction.

Usefulness of imputation for the analysis of incomplete otoneurologic data

The usefulness of imputation in the treatment of missing values of an otoneurologic database for the discriminant analysis was evaluated on the basis of the agreement of imputed values and the analysis results. The data consisted of six patient groups with vertigo (N=564). There were 38 variables and 11% of the data was missing. Missing values were filled in with the means, regression and Expectation-Maximisation (EM) imputation methods and a random imputation method provided the baseline results. Means, regression and EM methods agreed on 41-42% of the imputed missing values. The level of agreement between these and the random method was 20-22%. Despite the moderate agreement between the means, regression and EM methods, the discriminant functions were similar and accurate (prediction accuracy 83-99%). The discriminant functions obtained from the randomly imputed data were also accurate having prediction accuracy 88-97%. Imputation seems to be a useful method for treating the missing data in this database. However, a lot of data was missing in otoneurologic tests, which are likely to be of less importance in the diagnosis of vertiginous patients. Consequently, the disagreement of the methods did not affect clearly the discriminant analysis, and, therefore, future research requires more complete data and advanced imputation methods.