Artificial Intelligence for Real-Time Prediction of the Histology of Colorectal Polyps by General Endoscopists

BACKGROUND

Real-time prediction of histologic features of small colorectal polyps may prevent resection and/or pathologic evaluation and therefore decrease colonoscopy costs. Previous studies showed that computer-aided diagnosis (CADx) was highly accurate, though it did not outperform expert endoscopists.

OBJECTIVE

To assess the diagnostic performance of histologic predictions by general endoscopists before and after assistance from CADx in a real-life setting.

DESIGN

Prospective, multicenter, single-group study. (ClinicalTrials.gov: NCT04437615).

SETTING

6 centers across the United States.

PARTICIPANTS

1252 consecutive patients undergoing colonoscopy and 49 general endoscopists with variable experience in real-time prediction of polyp histologic features.

INTERVENTION

Real-time use of CADx during routine colonoscopy.

MEASUREMENTS

The primary end points were the sensitivity and specificity of CADx-unassisted and CADx-assisted histologic predictions for adenomas measuring 5 mm or less. For clinical purposes, additional estimates according to location and confidence level were provided.

RESULTS

The CADx device made a diagnosis for 2695 polyps measuring 5 mm or less (96%) in 1252 patients. There was no difference in sensitivity between the unassisted and assisted groups (90.7% vs. 90.8%; P = 0.52). Specificity was higher in the CADx-assisted group (59.5% vs. 64.7%; P < 0.001). Among all 2695 polyps measuring 5 mm or less, 88.2% and 86.1% (P < 0.001) in the CADx-assisted and unassisted groups, respectively, could be resected and discarded without pathologic evaluation. Among 743 rectosigmoid polyps measuring 5 mm or less, 49.5% and 47.9% (P < 0.001) in the CADx-assisted and unassisted groups, respectively, could be left in situ without resection.

LIMITATION

Decision making based on CADx might differ outside a clinical trial.

CONCLUSION

CADx assistance did not result in increased sensitivity of optical diagnosis. Despite a slight increase, the specificity of CADx-assisted diagnosis remained suboptimal.

PRIMARY FUNDING SOURCE

Olympus America Corporation served as the clinical study sponsor.

Optical diagnosis of colorectal polyps using novel blue light imaging classification among trainee endoscopists

BACKGROUND

Blue light imaging (BLI) has been shown to improve the characterization of colorectal polyps among the endoscopy experts. We aimed to determine if this technology could be taught to endoscopy trainees while maintaining high accuracy and interobserver agreement.

METHODS

Twenty-one gastroenterology trainees (fellows) from two academic institutions participated in this prospective study. Each trainee completed a web-based learning comprising four modules: pre-test, didactic videos explaining the BLI Adenoma Serrated International Classification (BASIC), interactive examples, and post-test assessment. The pre- and post-test modules consisted of reviewing video images of colon polyps in high definition white light imaging and BLI and then applying the BASIC classification to determine if the polyps were likely to be adenomatous. Confidence in adenoma identification (rated '1' to '5'), accuracy in polyp (adenoma vs. non-adenoma) identification, and agreement in characterization per BASIC criteria were derived.

RESULTS

Trainee accuracy in the adenoma diagnosis improved from 74.7% (pre-test) to 85.4% (post-test) (P < 0.01). There was a trend towards higher accuracy in polyp characterization with subsequent years of training (1st year fellows 77.4%, 2nd year 88.5%, and final year 94.0%) with consistent improvements after the e-learning across years of trainees. Overall, trainees were able to identify adenoma with a high sensitivity of 86.9%, specificity 83.9%, positive predictive value of 84.4%, and negative predictive value of 86.5%. However, their interobserver agreement in adenoma diagnosis was moderate (k = 0.52).

CONCLUSION

The novel BLI classification can be easily taught to gastroenterology trainees using an online module and accuracy improves with years of training reaching >90% for colorectal polyp characterization.

INTRATHECAL ADMINISTRATION OF HYPERBARIC MORPHINE FOR THE RELIEF OF PAIN IN LABOUR

Thirty healthy women in active labour received an intrathecal injection of morphine 0.5 mg (n = 12) or 1 mg (n = 18) in 7.5% dextrose. Both doses provided excellent analgesia for labour, 93% of patients obtaining at least 50% pain relief. Analgesia began 15-60 min after injection and did not decrease until 6-8 h after injection. Analgesia was satisfactory until distension of the perineum, either by forceps or the infant's head. The intrathecal injection of morphine did not adversely affect the condition of the infant. Eighty per cent of patients developed pruritus; 53%, nausea or vomiting, or both; 43%, urinary retention; and 43%, drowsiness. These side effects were decreased by naloxone, which did not affect the degree of analgesia. There was no significant depression of ventilation in any patient. These results suggest that morphine 0.5 mg or 1 mg, administered intrathecally, effectively decreases the pain of labour, and that i.v. administration of naloxone can alleviate the common side effects.

Continuous infusion epidural analgesia in parturients receiving bupivacaine, chloroprocaine, or lidocaine--maternal, fetal, and neonatal effects

The effects of epidural analgesia for labor and delivery using a continuous infusion technique on fetal heart rate, uterine activity, maternal blood pressure, Apgar scores, neonatal acid-base status, and the Neurologic and Adaptive Capacity Scoring System were studied in 61 parturients. Group I (n = 23) received initial test and therapeutic doses of 2 and 6 ml of 0.5% bupivacaine followed by an infusion of 0.125% at a rate of 14 ml/hr. Group II (n = 19) received 2 and 6 ml of 2% chloroprocaine followed by an infusion of 0.75% at a rate of 27 ml/hr. Group III (n = 19) received 2 and 6 ml of 1.5% lidocaine followed by an infusion of 0.75% at a rate of 14 ml/hr. None of the three local anesthetics used had any significant effect on baseline fetal heart rate or uterine activity. In cases in which monitoring of fetal heart rate was both technically satisfactory and continuous, late and variable decelerations in fetal heart rate were seen in 10 of 17, 3 of 18, and 2 of 19 of the fetuses in groups I, II, and III, respectively. The incidence was significantly higher in group I than in groups II or III (P less than 0.05). Apgar scores and neonatal acid-base status were equally good in all three groups. Neurologic and adaptive capacity scores did not differ among the three groups of neonates, nor did any of the neonates in the three groups score lower than a control group of 19 neonates whose mothers did not receive any analgesia or medications for labor and delivery.(ABSTRACT TRUNCATED AT 250 WORDS)

Epidural bupivacaine, chloroprocaine, or lidocaine for cesarean section--maternal and neonatal effects

The effects of epidural anesthesia on maternal blood pressure, newborn Apgar scores, neonatal acid-base status, and the early neonatal neurobehavioral scale (ENNS) were studied in 54 pregnant women undergoing cesarean section. Maternal and neonatal blood levels of the local anesthetics were also determined. Group 1 (n = 16) received 0.75% bupivacaine, group 2 (n = 18) received 3% chloroprocaine, group 3 (n = 11) received 2% lidocaine, and group 4 (n = 9) received 2% lidocaine with 1:200,000 epinephrine. The incidence of maternal hypotension did not differ significantly among the four groups. Fetal outcome as determined by Apgar scores, acid-base status, and neurobehavioral testing was equally good in all four groups. At delivery, the fetal/maternal concentration ratio of bupivacaine was 0.27, that of lidocaine without epinephrine 0.48, and that of lidocaine with epinephrine 0.58. Chloroprocaine was detected in 12 maternal samples, in seven umbilical venous samples, and in six umbilical arterial samples. It is concluded that epidural anesthesia as administered in this study had no adverse effect on the newborn Apgar scores, cord acid-base status, or the ENNS.

Effects of epidural anesthesia during labor on maternal plasma beta-endorphin levels

Plasma beta-endorphin (beta-EP) was measured in 48 women. Twenty-three were in labor. In 13 of the 23 patients in labor, beta-EP was determined prior to and after complete onset of epidural anesthesia, and in 10 women, who served as controls, prior to and after injection of saline into the epidural space as part of the loss of resistance technique, but before injection of the local anesthetic. Venous blood also was obtained for plasma beta-EP determinations from 10 healthy non-pregnant women and from 15 patients scheduled for elective repeat cesarean section and who were not in labor. Human beta-EP was determined by radioimmunoassay following silicic acid extraction of plasma samples and separation of the beta-EP fraction by gel chromatography. In the 10 non-pregnant volunteers, plasma beta-EP averaged 11.3 +/- 1.5 fmol/ml (mean +/- SE) as compared with 43.7 +/- 6.5 fmol/ml observed in the 15 women with term pregnancies who were not in labor (P less than 0.005). In the 13 patients in labor who underwent epidural anesthesia, plasma beta-EP concentrations decreased (P less than 0.005) from 54.5 +/- 9.0 to 28.2 +/- 3.5 fmol/ml, whereas there was no significant change in plasma beta-EP levels in the 10 controls who averaged 64 +/- 20.5 and 55.8 +/- 13.6 fmol/ml prior to and following saline injection. These data confirm that plasma beta-EP levels are significantly higher in women with term pregnancies in labor than in non-pregnant women and also demonstrate that epidural anesthesia during labor is accompanied by a significant decrease in maternal plasma beta-EP concentrations.

Lack of adverse neonatal neurobehavioral effects of lidocaine

The effects of epidural anesthesia on the early neonatal neurobehavioral scale (ENNS) following delivery were studied in 22 parturients given 1.5% lidocaine epidural anesthesia during labor and delivery. These results were compared to those obtained in a control group of 17 neonates whose mothers received no analgesics, medications, or local anesthetics for labor or delivery. There was no significant difference in ENNS in the two groups of neonates. Lidocaine as administered in this study has no adverse effects on the early neurobehavioral status of the neonate.

Sympathoadrenal activity, maternal, fetal, and neonatal responses after epidural anesthesia in the preeclamptic patient

This study confirms that preeclamptic patients have higher plasma levels of catecholamine than those of normal patients. It also demonstrates that epidural analgesia when administered to the preeclamptic patient during labor is followed by a significant reduction in the plasma levels of catecholamines without any adverse effects on maternal blood pressure, uterine activity, fetal heart rate, or the neonate.