-
Clinical Journal of the American... Mar 2008Acute kidney injury (previously known as acute renal failure) is a common complication in hospitalized patients, and its incidence has risen significantly in the past 15... (Review)
Review
Acute kidney injury (previously known as acute renal failure) is a common complication in hospitalized patients, and its incidence has risen significantly in the past 15 yr. Despite significant technical advances in therapeutics, the mortality and morbidity rates associated with acute kidney injury remain dismally high and have not appreciably improved during the past four decades. Although the serum creatinine concentration performs fairly well for estimating kidney function in patients with stable chronic kidney disease, it performs poorly in the setting of acute disease. An ideal biomarker for acute kidney injury would help clinicians and scientists diagnose the most common form of acute kidney injury in hospitalized patients, acute tubular necrosis, early and accurately and may aid to risk-stratify patients with acute kidney injury by predicting the need for renal replacement therapy, the duration of acute kidney injury, the length of stay, and mortality. Herein is reviewed the diagnostic and prognostic performance of several types of urinary biomarkers for the diagnosis and risk stratification of acute kidney injury. The major types of urinary biomarkers fall into three classes: (1) Inflammatory, (2) renal tubular proteins that are excreted into the urine after injury, and (3) surrogate markers of tubular injury. Also discussed are statistical issues in evaluating the accuracy of biomarkers as diagnostic tests. It is likely that a panel of biomarkers, rather than a single biomarker, will be needed to perform extremely well in these three situations.
Topics: Acute Kidney Injury; Biomarkers; Diagnosis, Differential; Early Diagnosis; Humans; Reproducibility of Results; Risk Assessment
PubMed: 18256377
DOI: 10.2215/CJN.03520807 -
Fukushima Journal of Medical Science Aug 2017Endoscopic ultrasonography (EUS) plays a major role in diagnosing gallbladder (GB) cancer and pancreatic cancer (PC). In cases of GB cancer, EUS allows for precise... (Review)
Review
Endoscopic ultrasonography (EUS) plays a major role in diagnosing gallbladder (GB) cancer and pancreatic cancer (PC). In cases of GB cancer, EUS allows for precise observations of morphology and wall layers. However, proficiency is required for the morphologic diagnosis of GB tumors. Therefore, contrast-enhanced harmonic EUS (CH-EUS) began to be performed to diagnose GB lesions. CH-EUS enables real-time observation of the hemodynamics of GB tumors. The enhanced patterns generated by CH-EUS improve precision in the diagnosis of such tumors.PC appears as a hypoechoic mass on EUS. However, distinguishing between PC and mass-forming pancreatitis or focal autoimmune pancreatitis (AIP) is difficult via conventional EUS. CH-EUS allows for differentiating among these diseases (PC is hypoenhanced and heterogeneously enhanced, pancreatitis is isoenhanced, and a pancreatic neuroendocrine tumor is hyperenhanced). EUS-guided fine needle aspiration (EUS-FNA) also contributes to pathological diagnoses of pancreatic lesions. However, certain PC patients cannot be diagnosed via EUS-FNA. PC is heterogeneously enhanced on CH-EUS, and unenhanced regions have been reported to be areas of fibrosis or necrosis. CH-EUS-guided fine needle aspiration (CH-EUS-FNA) permits puncturing of the enhanced area while avoiding necrotic and fibrotic regions. Moreover, as CH-EUS findings have been quantitatively analyzed, a time-intensity curve (TIC) has become usable for diagnosing solid pancreatic lesions. CH-EUS-related techniques have been developed and increasingly utilized in the pancreaticobiliary area.
Topics: Contrast Media; Endoscopic Ultrasound-Guided Fine Needle Aspiration; Endosonography; Gallbladder Neoplasms; Humans; Image Enhancement; Pancreatic Neoplasms
PubMed: 28680009
DOI: 10.5387/fms.2017-04 -
BMC Endocrine Disorders Jun 2023To compare the ability of the Cox regression and machine learning algorithms to predict the survival of patients with Anaplastic thyroid carcinoma (ATC).
BACKGROUND
To compare the ability of the Cox regression and machine learning algorithms to predict the survival of patients with Anaplastic thyroid carcinoma (ATC).
METHODS
Patients diagnosed with ATC were extracted from the Surveillance, Epidemiology, and End Results database. The outcomes were overall survival (OS) and cancer-specific survival (CSS), divided into: (1) binary data: survival or not at 6 months and 1 year; (2): time-to-event data. The Cox regression method and machine learnings were used to construct models. Model performance was evaluated using the concordance index (C-index), brier score and calibration curves. The SHapley Additive exPlanations (SHAP) method was deployed to interpret the results of machine learning models.
RESULTS
For binary outcomes, the Logistic algorithm performed best in the prediction of 6-month OS, 12-month OS, 6-month CSS, and 12-month CSS (C-index = 0.790, 0.811, 0.775, 0.768). For time-event outcomes, traditional Cox regression exhibited good performances (OS: C-index = 0.713; CSS: C-index = 0.712). The DeepSurv algorithm performed the best in the training set (OS: C-index = 0.945; CSS: C-index = 0.834) but performs poorly in the verification set (OS: C-index = 0.658; CSS: C-index = 0.676). The brier score and calibration curve showed favorable consistency between the predicted and actual survival. The SHAP values was deployed to explain the best machine learning prediction model.
CONCLUSIONS
Cox regression and machine learning models combined with the SHAP method can predict the prognosis of ATC patients in clinical practice. However, due to the small sample size and lack of external validation, our findings should be interpreted with caution.
Topics: Humans; Thyroid Carcinoma, Anaplastic; Algorithms; Databases, Factual; Machine Learning; Thyroid Neoplasms; Prognosis
PubMed: 37291551
DOI: 10.1186/s12902-023-01368-5 -
Medical Decision Making : An... May 2016The unconscious thought theory argues that making complex decisions after a period of distraction can lead to better decision quality than deciding either immediately or...
The unconscious thought theory argues that making complex decisions after a period of distraction can lead to better decision quality than deciding either immediately or after conscious deliberation. Two studies have tested this unconscious thought effect (UTE) in clinical diagnosis with conflicting results. The studies used different methodologies and had methodological weaknesses. We attempted to replicate the UTE in medical diagnosis by providing favorable conditions for the effect while maintaining ecological validity. Family physicians (N= 116) diagnosed 3 complex cases in 1 of 3 thinking modes: immediate, unconscious (UT), and conscious (CT). Cases were divided into short sentences, which were presented briefly and sequentially on computer. After each case presentation, the immediate response group gave a diagnosis, the UT group performed a 2-back distraction task for 3 min before giving a diagnosis, and the CT group could take as long as necessary before giving a diagnosis. We found no differences in diagnostic accuracy between groups (P= 0.95). The CT group took a median of 7 s to diagnose, which suggests that physicians were able to diagnose "online," as information was being presented. The lack of a difference between the immediate and UT groups suggests that the distraction had no additional effect on performance. To assess the decisiveness of the evidence of this null result, we computed a Bayes factor (BF01) for the 2 comparisons of interest. We found a BF01of 5.76 for the UT versus immediate comparison and of 3.61 for the UT versus CT comparison. Both BFs provide substantial evidence in favor of the null hypothesis: physicians' diagnoses made after distraction are no better than diagnoses made either immediately or after self-paced deliberation.
Topics: Adult; Aged; Bayes Theorem; Clinical Decision-Making; Diagnosis; Female; Humans; Male; Middle Aged; Models, Psychological; Physicians, Family; Reproducibility of Results; Unconscious, Psychology
PubMed: 25852079
DOI: 10.1177/0272989X15581352 -
Scientific Reports Jul 2022Computed tomography (CT) has been widely used to diagnose Graves' orbitopathy, and the utility is gradually increasing. To develop a neural network (NN)-based method for...
Computed tomography (CT) has been widely used to diagnose Graves' orbitopathy, and the utility is gradually increasing. To develop a neural network (NN)-based method for diagnosis and severity assessment of Graves' orbitopathy (GO) using orbital CT, a specific type of NN optimized for diagnosing GO was developed and trained using 288 orbital CT scans obtained from patients with mild and moderate-to-severe GO and normal controls. The developed NN was compared with three conventional NNs [GoogleNet Inception v1 (GoogLeNet), 50-layer Deep Residual Learning (ResNet-50), and 16-layer Very Deep Convolutional Network from Visual Geometry group (VGG-16)]. The diagnostic performance was also compared with that of three oculoplastic specialists. The developed NN had an area under receiver operating curve (AUC) of 0.979 for diagnosing patients with moderate-to-severe GO. Receiver operating curve (ROC) analysis yielded AUCs of 0.827 for GoogLeNet, 0.611 for ResNet-50, 0.540 for VGG-16, and 0.975 for the oculoplastic specialists for diagnosing moderate-to-severe GO. For the diagnosis of mild GO, the developed NN yielded an AUC of 0.895, which is better than the performances of the other NNs and oculoplastic specialists. This study may contribute to NN-based interpretation of orbital CTs for diagnosing various orbital diseases.
Topics: Graves Ophthalmopathy; Humans; Neural Networks, Computer; Tomography, X-Ray Computed
PubMed: 35840769
DOI: 10.1038/s41598-022-16217-z -
International Journal of Stroke :... Feb 2019Identifying and treating patients with transient ischemic attack is an effective means of preventing stroke. However, making this diagnosis can be challenging, and over...
BACKGROUND
Identifying and treating patients with transient ischemic attack is an effective means of preventing stroke. However, making this diagnosis can be challenging, and over a third of patients referred to stroke prevention clinic are ultimately found to have alternate diagnoses.
AIMS
We performed a systematic review to determine how neurologists diagnose transient ischemic attack.
SUMMARY OF REVIEW
A systematic literature search was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines using MEDLINE, Embase, and the Cochrane Library databases. Publications eligible for inclusion were those that included information on the demographic or clinical features neurologists use to diagnose transient ischemic attacks or transient ischemic attack-mimics. Of 1666 citations, 210 abstracts were selected for full-text screening and 80 publications were ultimately deemed eligible for inclusion. Neurologists were more likely to diagnose transient ischemic attack based on clinical features including negative symptoms or speech deficits. Patients with positive symptoms, altered level of consciousness, or the presence of nonfocal symptoms such as confusion or amnesia were more likely to be diagnosed with transient ischemic attack-mimic. Neurologists commonly include mode of onset (i.e. sudden versus gradual), recurrence of attacks, and localizability of symptoms to a distinct vascular territory in the diagnostic decision-making process. Transient ischemic attack diagnosis was more commonly associated with advanced age, preexisting hypertension, atrial fibrillation, and other vascular risk factors.
CONCLUSIONS
Neurologists rely on certain clinical and demographic features to distinguish transient ischemic attacks from mimics, which are not currently reflected in widely used risk scores. Clarifying how neurologists diagnose transient ischemic attack may help frontline clinicians to better select patients for referral to stroke prevention clinics.
Topics: Atrial Fibrillation; Canada; Clinical Decision-Making; Diagnosis, Differential; Humans; Hypertension; Ischemic Attack, Transient; Neurologists; Risk Factors; Stroke
PubMed: 30507363
DOI: 10.1177/1747493018816430 -
The Tohoku Journal of Experimental... Dec 2022Imaging features of the lung in postmortem computed tomography (CT) scans have been reported in drowning cases. However, it is difficult for forensic pathologists with...
Imaging features of the lung in postmortem computed tomography (CT) scans have been reported in drowning cases. However, it is difficult for forensic pathologists with limited experience to distinguish subtle differences in CT images. In this study, artificial intelligence (AI) with deep learning capability was used to diagnose drowning in postmortem CT images, and its performance was evaluated. The samples consisted of high-resolution CT images of the chest of 153 drowned and 160 non-drowned bodies captured by an 8- or 64-row multislice CT system. The images were captured with an image slice thickness of 1.0 mm and spacing of 30 mm, and 28 images were typically captured. A modified AlexNet was used as the AI architecture. The output result was the drowning probability for each component image. To evaluate the performance of the proposed model, the area under the receiver operating characteristic curve (AUC) was analyzed, and the AUC value of 0.95 was obtained. This indicates that the proposed AI architecture is a useful and powerful complementary testing approach for diagnosing drowning in postmortem CT images. Notably, the accuracy was 81% (62/77) for cases in which resuscitation was performed, and 92% (216/236) for cases in which resuscitation was not attempted. Therefore, the proposed AI method should not be used to diagnose the cause of death when aggressive cardiopulmonary resuscitation was performed. Additionally, because honeycomb lungs are likely to exhibit different morphologies, emphysema cases should also be treated with caution when the proposed AI method is used to diagnose drowning.
Topics: Humans; Drowning; Artificial Intelligence; Tomography, X-Ray Computed; Lung; ROC Curve
PubMed: 36384859
DOI: 10.1620/tjem.2022.J097 -
World Journal of Surgical Oncology Jan 2019In this study, images of 2450 benign thyroid nodules and 2557 malignant thyroid nodules were collected and labeled, and an automatic image recognition and diagnosis...
BACKGROUND
In this study, images of 2450 benign thyroid nodules and 2557 malignant thyroid nodules were collected and labeled, and an automatic image recognition and diagnosis system was established by deep learning using the YOLOv2 neural network. The performance of the system in the diagnosis of thyroid nodules was evaluated, and the application value of artificial intelligence in clinical practice was investigated.
METHODS
The ultrasound images of 276 patients were retrospectively selected. The diagnoses of the radiologists were determined according to the Thyroid Imaging Reporting and Data System; the images were automatically recognized and diagnosed by the established artificial intelligence system. Pathological diagnosis was the gold standard for the final diagnosis. The performances of the established system and the radiologists in diagnosing the benign and malignant thyroid nodules were compared.
RESULTS
The artificial intelligence diagnosis system correctly identified the lesion area, with an area under the receiver operating characteristic (ROC) curve of 0.902, which is higher than that of the radiologists (0.859). This finding indicates a higher diagnostic accuracy (p = 0.0434). The sensitivity, positive predictive value, negative predictive value, and accuracy of the artificial intelligence diagnosis system for the diagnosis of malignant thyroid nodules were 90.5%, 95.22%, 80.99%, and 90.31%, respectively, and the performance did not significantly differ from that of the radiologists (p > 0.05). The artificial intelligence diagnosis system had a higher specificity (89.91% vs 77.98%, p = 0.026).
CONCLUSIONS
Compared with the performance of experienced radiologists, the artificial intelligence system has comparable sensitivity and accuracy for the diagnosis of malignant thyroid nodules and better diagnostic ability for benign thyroid nodules. As an auxiliary tool, this artificial intelligence diagnosis system can provide radiologists with sufficient assistance in the diagnosis of benign and malignant thyroid nodules.
Topics: Adult; Aged; Diagnosis, Differential; Female; Humans; Image Interpretation, Computer-Assisted; Male; Middle Aged; Neural Networks, Computer; Predictive Value of Tests; Prognosis; ROC Curve; Retrospective Studies; Thyroid Gland; Thyroid Nodule; Ultrasonography; Young Adult
PubMed: 30621704
DOI: 10.1186/s12957-019-1558-z -
Urology Journal 2011To determine the value of color Doppler ultrasonography (CDUS) as a routine investigational method for diagnosis of scrotal pathologies. (Review)
Review
PURPOSE
To determine the value of color Doppler ultrasonography (CDUS) as a routine investigational method for diagnosis of scrotal pathologies.
MATERIALS AND METHODS
This prospective observational study (case series) was carried out over a period of 16 months on 122 patients in the age range of 13 to 70 years old, who presented with scrotal swellings. After adequate history taking and examination, CDUS was performed. The diagnosis of the surgeon and that of radiologist were compared with final outcome, which was based on course and outcome of the disease, fine needle aspiration cytology results, and operative findings.
RESULTS
The final diagnoses were epididymitis or epididymo-orchitis (46), hydrocele (26), varicocele (16), testicular malignancy (16), orchitis (6), testicular torsion (4), spermatic cord injury (2), hematocele (2), and pyocele (2). Color Doppler ultrasonography accurately diagnosed all cases of epididymitis or epididymo-orchitis, spermatic cord injury, testicular torsion, varicocele, and hydrocele (sensitivity 100% and specificity 100%). Of 16 subjects diagnosed as testicular malignancy on CDUS, only 14 were subsequently found to have malignancy. Two cases of orchitis were wrongly diagnosed as malignancy. Similarly, of 6 patients diagnosed as orchitis, 1 was found to have seminoma (sensitivity 87.5% and specificity 66.7%). Overall sensitivity of CDUS in diagnosing scrotal diseases was 98% while specificity was 66.7%.
CONCLUSION
Color Doppler ultrasonography is an excellent, a safe, and reliable method for evaluating patients with scrotal diseases. It aids in diagnosis of testicular tumors and reduces the number of unnecessary exploratory operations. It is especially important in conditions like testicular torsion where immediate diagnosis is required.
Topics: Adolescent; Adult; Aged; Genital Diseases, Male; Humans; Male; Middle Aged; Prospective Studies; Scrotum; Ultrasonography, Doppler, Color; Young Adult
PubMed: 21404205
DOI: No ID Found -
Sensors (Basel, Switzerland) Nov 2022Computer-aided diagnosis (CAD) has proved to be an effective and accurate method for diagnostic prediction over the years. This article focuses on the development of an...
Computer-aided diagnosis (CAD) has proved to be an effective and accurate method for diagnostic prediction over the years. This article focuses on the development of an automated CAD system with the intent to perform diagnosis as accurately as possible. Deep learning methods have been able to produce impressive results on medical image datasets. This study employs deep learning methods in conjunction with meta-heuristic algorithms and supervised machine-learning algorithms to perform an accurate diagnosis. Pre-trained convolutional neural networks (CNNs) or auto-encoder are used for feature extraction, whereas feature selection is performed using an ant colony optimization (ACO) algorithm. Ant colony optimization helps to search for the best optimal features while reducing the amount of data. Lastly, diagnosis prediction (classification) is achieved using learnable classifiers. The novel framework for the extraction and selection of features is based on deep learning, auto-encoder, and ACO. The performance of the proposed approach is evaluated using two medical image datasets: chest X-ray (CXR) and magnetic resonance imaging (MRI) for the prediction of the existence of COVID-19 and brain tumors. Accuracy is used as the main measure to compare the performance of the proposed approach with existing state-of-the-art methods. The proposed system achieves an average accuracy of 99.61% and 99.18%, outperforming all other methods in diagnosing the presence of COVID-19 and brain tumors, respectively. Based on the achieved results, it can be claimed that physicians or radiologists can confidently utilize the proposed approach for diagnosing COVID-19 patients and patients with specific brain tumors.
Topics: Humans; COVID-19; Deep Learning; Diagnosis, Computer-Assisted; Brain Neoplasms; Computers
PubMed: 36433595
DOI: 10.3390/s22228999