Artificial intelligence, BI-RADS evaluation and morphometry: A novel combination to diagnose breast cancer using ultrasonography, results from multi-center cohorts.

PURPOSE: To develop and validate a machine learning (ML) model for the classification of breast lesions on ultrasound images. METHOD: In the present study, three separate data cohorts containing 1288 breast lesions from three countries (Malaysia, Iran, and Turkey) were utilized for MLmodel development and external validation. The model was trained on ultrasound images of 725 breast lesions, and validation was done separately on the remaining data. An expert radiologist and a radiology resident classified the lesions based on the BI-RADS lexicon. Thirteen morphometric features were selected from a contour of the lesion and underwent a three-step feature selection process. Five features were chosen to be fed into the model separately and combined with the imaging signs mentioned in the BI-RADS reference guide. A support vector classifier was trained and optimized. RESULTS: The diagnostic profile of the model with various input data was compared to the expert radiologist and radiology resident. The agreement of each approach with histopathologic specimens was also determined. Based on BI-RADS and morphometric features, the model achieved an area under the receiver operating characteristic (ROC) curve (AUC) of 0.885, which is higher than the expert radiologist and radiology resident performances with AUC of 0.814 and 0.632, respectively in all cohorts. DeLong's test also showed that the AUC of the ML protocol was significantly different from that of the expert radiologist (ΔAUCs = 0.071, 95%CI: (0.056, 0.086), P = 0.005). CONCLUSIONS: These results support the possible role of morphometric features in enhancing the already well-excepted classification schemes.

Comparing point shear wave elastography (ElastPQ) and transient elastography for diagnosis of fibrosis stage in non-alcoholic fatty liver disease.

BACKGROUND AND AIM: Transient elastography (TE) and point shear wave elastography (pSWE) are noninvasive methods to diagnose fibrosis stage in patients with chronic liver disease. The aim of this study is to compare the accuracy of the two methods to diagnose fibrosis stage in non-alcoholic fatty liver disease (NAFLD) and to study the intra-observer and inter-observer variability when the examinations were performed by healthcare personnel of different backgrounds. METHODS: Consecutive NAFLD patients who underwent liver biopsy were enrolled in this study and had two sets each of pSWE and TE examinations by a nurse and a doctor on the same day of liver biopsy procedure. The medians of the four sets of pSWE and TE were used for evaluation of diagnostic accuracy using area under receiver operating characteristic curve (AUROC). Intra-observer and inter-observer variability was analyzed using intraclass correlation coefficients. RESULTS: Data for 100 NAFLD patients (mean age 57.1 ± 10.2 years; male 46.0%) were analyzed. The AUROC of TE for diagnosis of fibrosis stage ≥ F1, ≥ F2, ≥ F3, and F4 was 0.89, 0.83, 0.83, and 0.89, respectively. The corresponding AUROC of pSWE was 0.80, 0.72, 0.69, and 0.79, respectively. TE was significantly better than pSWE for the diagnosis of fibrosis stages ≥ F2 and ≥ F3. The intra-observer and inter-observer variability of TE and pSWE measurements by the nurse and doctor was excellent with intraclass correlation coefficient > 0.96. CONCLUSION: Transient elastography was significantly better than pSWE for the diagnosis of fibrosis stage ≥ F2 and ≥ F3. Both TE and pSWE had excellent intra-observer and inter-observer variability when performed by healthcare personnel of different backgrounds.

Automated detection and classification of liver fibrosis stages using contourlet transform and nonlinear features.

BACKGROUND AND OBJECTIVE: Liver fibrosis is a type of chronic liver injury that is characterized by an excessive deposition of extracellular matrix protein. Early detection of liver fibrosis may prevent further growth toward liver cirrhosis and hepatocellular carcinoma. In the past, the only method to assess liver fibrosis was through biopsy, but this examination is invasive, expensive, prone to sampling errors, and may cause complications such as bleeding. Ultrasound-based elastography is a promising tool to measure tissue elasticity in real time; however, this technology requires an upgrade of the ultrasound system and software. In this study, a novel computer-aided diagnosis tool is proposed to automatically detect and classify the various stages of liver fibrosis based upon conventional B-mode ultrasound images. METHODS: The proposed method uses a 2D contourlet transform and a set of texture features that are efficiently extracted from the transformed image. Then, the combination of a kernel discriminant analysis (KDA)-based feature reduction technique and analysis of variance (ANOVA)-based feature ranking technique was used, and the images were then classified into various stages of liver fibrosis. RESULTS: Our 2D contourlet transform and texture feature analysis approach achieved a 91.46% accuracy using only four features input to the probabilistic neural network classifier, to classify the five stages of liver fibrosis. It also achieved a 92.16% sensitivity and 88.92% specificity for the same model. The evaluation was done on a database of 762 ultrasound images belonging to five different stages of liver fibrosis. CONCLUSIONS: The findings suggest that the proposed method can be useful to automatically detect and classify liver fibrosis, which would greatly assist clinicians in making an accurate diagnosis.

Impact of operator fatigue on endoscopy performance: implications for procedure scheduling.

BACKGROUND: With increasing volumes of endoscopic procedures, endoscopists' workload has had to increase to meet this escalating demand. The aim of this study was to characterize the impact of endoscopist fatigue on quality of endoscopy performance by comparing outcomes based on chronological procedure order. METHODS: Consecutive endoscopic procedures were prospectively observed. Quality indicators of colonoscopy (cecal intubation rate, lesion detection, withdrawal time, insertion time) and esophagogastroduodenoscopy (EGD) duration were compared among procedures based on their chronological sequence. RESULTS: Colonoscopy completion rates declined with successive procedures; completion for 1st to 3rd procedures (90%) was significantly higher than for 4th and subsequent procedures (76%) (P = 0.03). Median insertion times lengthened; times for 1st to 4th procedures [8 min, interquartile range (IQR) 6-11 min] were shorter than for 5th and subsequent procedures (10 min, IQR 7-15 min) (P = 0.06). Lesion detection rates, withdrawal times, and EGD duration remained stable with procedure order. CONCLUSIONS: Colonoscopy cecal intubation rates appear to decline with successive procedures. There also appears to be a trend for insertion times to lengthen. Reassuringly, other quality indicators of colonoscopy (lesion detection and withdrawal time) and EGD duration do not appear to be impacted by repetitive procedures.

A "time-and-motion" study of endoscopic practice: strategies to enhance efficiency.

BACKGROUND: With the growing demand on endoscopic resources, achieving optimal efficiency has assumed increasing importance. OBJECTIVE: This study adopted a time-and-motion approach to assess efficiency in the endoscopy unit of a large teaching hospital and to identify strategies to enhance efficiency. DESIGN: Consecutive endoscopic procedures were prospectively observed over the study period, and time intervals of the individual components of each procedure were recorded. SETTING: Tertiary-referral teaching hospital. PATIENTS: Consecutive patients undergoing endoscopy. INTERVENTION: Prospective recording of endoscopic data. MAIN OUTCOME MEASUREMENTS: Time intervals of the individual components of each procedure. RESULTS: Data were prospectively recorded for 400 procedures: 197 EGDs, 123 colonoscopies, 32 flexible sigmoidoscopies, and 48 double procedures (an EGD and a flexible sigmoidoscopy or colonoscopy). Several strategies to improve the efficiency quotient (EQ), the proportion of time that the endoscopist is engaged in performing the procedure or completing postprocedure paperwork, were identified: (1) employing personnel to obtain prior intravenous access and consent of patients increased the EQ by 10.8%, (2) using a 2-rooms-per-endoscopist model increased the EQ by 51.2%, (3) using personnel to both obtain consent and sedate the patient before an endoscopy increased the EQ by 30.9%, and (4) eliminating postprocedure paperwork for the endoscopist in conjunction with preconsent and sedation and a 2-room model increased the EQ by 63.3%. LIMITATIONS: Findings represent the experience of a single endoscopy unit in a tertiary-referral center and may not be generalizable to ambulatory surgical centers or other hospital-based endoscopy units. Factors other than procedure-time components may impact the efficiency of a 2-rooms-per-endoscopist model. CONCLUSIONS: A time-and-motion approach can be used to identify strategies to enhance endoscopic efficiency. The quality of any aspect of endoscopy performance should never be compromised in an attempt to enhance efficiency.