Search

ImportanceTelemedicine use accelerated during the COVID-19 pandemic, and skin conditions were a common use case. However, many images submitted may be of insufficient quality for making a clinical determination.ObjectiveTo determine whether an artificial intelligence (AI) decision support tool, a machine learning algorithm, could improve the quality of images submitted for telemedicine by providing real-time feedback and explanations to patients.Design, Setting, and ParticipantsThis quality improvement study with an AI performance component and single-arm clinical pilot study component was conducted from March 2020 to October 2021. After training, the AI decision support tool was tested on 357 retrospectively collected telemedicine images from Stanford telemedicine from March 2020 to June 2021. Subsequently, a single-arm clinical pilot study was conducted to assess feasibility with 98 patients in the Stanford Department of Dermatology across 2 clinical sites from July 2021 to October 2021. For the clinical pilot study, inclusion criteria for patients included being adults (aged ≥18 years), presenting to clinic for a skin condition, and being able to photograph their own skin with a smartphone.InterventionsDuring the clinical pilot study, patients were given a handheld smartphone device with a machine learning algorithm interface loaded and were asked to take images of any lesions of concern. Patients were able to review and retake photos prior to submitting, so each submitted photo met the patient’s assumed standard of clinical acceptability. A machine learning algorithm then gave the patient feedback on whether the image was acceptable. If the image was rejected, the patient was provided a reason by the AI decision support tool and allowed to retake the photos.Main Outcomes and MeasuresThe main outcome of the retrospective image analysis was the receiver operator curve area under the curve (ROC-AUC). The main outcome of the clinical pilot study was the image quality difference between the baseline images and the images approved by AI decision support.ResultsOf the 98 patients included, the mean (SD) age was 49.8 (17.6) years, and 50 (51%) of the patients were male. On retrospective telemedicine images, the machine learning algorithm effectively identified poor-quality images (ROC-AUC of 0.78) and the reason for poor quality (blurry ROC-AUC of 0.84; lighting issues ROC-AUC of 0.70). The performance was consistent across age and sex. In the clinical pilot study, patient use of the machine learning algorithm was associated with improved image quality. An AI algorithm was associated with reduction in the number of patients with a poor-quality image by 68.0%.Conclusions and RelevanceIn this quality improvement study, patients use of the AI decision support with a machine learning algorithm was associated with improved quality of skin disease photographs submitted for telemedicine use.

Atypical fibroxanthoma and undifferentiated pleomorphic sarcoma, or pleomorphic dermal sarcoma, are rare malignant cutaneous neoplasms existing along a clinicopathologic spectrum. Although these tumors share many similarities, recognition of distinguishing characteristics may predict differences in clinical behavior and outcomes. Salient features defining atypical fibroxanthoma include superficial tumors with minimal high-risk histologic features. Deeper tumors with high-risk histologic features are often clinically aggressive and should be appropriately designated as pleomorphic dermal sarcoma. Surgery remains gold standard in management; tumor extirpation with complete margin control is critical. In the high-risk tumor cohort, comprehensive evaluation and multidisciplinary management is paramount for optimal outcomes.

We describe a hydrodynamic simulation method (HSM) that is based on hydrodynamic considerations in Batch and Semi Batch stirred reactor systems. The method combines hydrodynamic studies of the mixing procedure obtained from experiments in small and large scale stirred reactors together with process simulation by VisiMix software. We describe how this hydrodynamic simulation method can aid in process optimization and scale up. The use of the simulation method described in this article, will offer the user the possibility to achieve the best results during production stage, saving time and currency, and at the same time increasing the knowledge of the performed process. Several examples in the article demonstrate the benefits of the proposed method.

Does coronary artery reoperation improve the survival of patients with stenoses in saphenous vein bypass grafts to coronary arteries? To examine this question, we retrospectively reviewed 1117 patients who had coronary bypass grafting and then underwent a postoperative coronary angiogram that showed a stenosis (≥20%) of at least one vein graft. Reoperation within 1 month of the postoperative angiogram was performed for 394 patients (REOP group) whereas 723 patients MED group) received initial medical treatment (no reoperation or percutaneous transluminal coronary angioplasty within 1 year). Compared with the MED group, patients in the REOP group were older, more symptomatic, more likely to have left main stenosis, and had fewer patent bypass grafts (all p

From August 1979 through August 1991, 91 patients were supported with centrifugal mechanical ventricular assist. Major indications for its use were postcardiotomy ventricular failure (79) or as a bridge to cardiac transplantation (12). In postcardiotomy use (0.2% of all cardiac procedures), there were 54 male (68.4%) and 25 female patients (31.6%) with a mean age of 54.8 years and a mean duration of use of 3.56 days (range, 1 hour to 19 days). Forty-nine patients (62%) were successfully weaned, and 20 (25.3%) were hospital survivors. In 57 patients the device was inserted to wean from cardiopulmonary bypass, whereas in 22 it was employed later in the postoperative period because of low cardiac output or sudden arrest. Thirty-four (59.6%) of the 57 patients in the former group were weaned, and 15 (26.3%) were discharged, results similar to those in the latter group with 15 (68.2%) weaned and 5 (22.7%) discharged. Morbidity associated with use of centrifugal blood pumps included bleeding (87.3%; mean transfusion requirement, 53.2 units), renal failure (46.8%), cerebrovascular accident (12.7%), thromboembolism (12.7%), and hepatic insufficiency (12.7%). After a mean follow-up of 45.4 months (range, 2 to 142 months), 7 patients had died (35% late mortality), 1 patient is in functional class IV, and all others are in functional class I or II. Lower survival was associated with biventricular failure and renal failure but not with age or sex of the patient.

The belief that postoperative atrial fibrillation (PAF) results from transient autonomic dysfunction suggests that interventions such as clinical hypnosis may reduce the incidence of PAF. To explore this hypothesis, we retrospectively compared outcomes between two groups of patients undergoing coronary artery bypass graft surgery: 50 consecutive patients who received preoperative hypnoidal explanation of the surgical procedure and 50 case-matched historical controls who received no clinical hypnosis. The patients who received hypnosis were significantly less likely to experience an episode of PAF (P = .003) and showed nonsignificant trends toward superior outcomes in terms of length of stay, narcotic use, and total hospital charges. Our findings indicate that prospective randomized trials are warranted to further delineate the potential benefit of clinical hypnosis for prevention of PAF.

We tested the efficacy and safety of aprotinin in 169 patients undergoing isolated reoperative myocardial revascularization. Patients were randomly assigned to high-dose aprotinin, low-dose aprotinin, or placebo treatment groups in a double-blind, placebo-controlled study. Treatment groups did not differ significantly with respect to age, sex, red cell mass, number of grafts, use of internal thoracic artery, or incidence of preoperative aspirin therapy. Patients treated with aprotinin had a significant reduction in postoperative chest tube drainage (720 +/- 753, 866 +/- 1,636, and 1,121 +/- 683 mL, respectively, for high-dose aprotinin, low-dose aprotinin, and placebo; p0.001). Transfusion requirements were reduced in aprotinin-treated patients (2.1 +/- 4.2, 4.8 +/- 11.8, and 4.1 +/- 6.2 units for high-dose, low-dose, and placebo, respectively; p0.001). A similar reduction in chest tube drainage and transfusion requirements was seen in patients using aspirin preoperatively. Q-wave myocardial infarctions were increased in the aprotinin subgroups (17.5%, 14.3%, and 8.9% for high-dose, low-dose, and placebo groups; not significant). Acute vein graft thrombosis was found in six of 12 vein grafts studied at postmortem examination in patients receiving aprotinin but not in any of five grafts in patients receiving placebo. We conclude that aprotinin is extremely effective in reducing bleeding and transfusion requirements and may increase the risk of graft thrombosis.

When systemic cooling and rewarming are performed during cardiopulmonary bypass (CPB), the pulmonary artery temperature typically decreases after CPB. This decrease may be rapid enough to cause substantial underestimation of cardiac output (CO) measured by thermodilution, due to changing baseline temperature during the thermodilution measurement. In 16 patients undergoing CPB for coronary artery grafts, digital recording of pulmonary artery temperature was done during room-temperature thermodilution CO (TDCO) injections. TDCO were computed with and without correction for baseline temperature decrease. Prior to CPB, the temperature change was -0.013 degrees C/min, producing no significant effect on CO measurements; the coefficient of variation of CO measurements was 5.1%. One minute after CPB the temperature change was -0.144 degrees C/min, producing a CO measurement error of -0.57 +/- 0.52 l/min (SD), or about 11% of the average CO; the range of the error was 0.05 to -2.0 l/min. Ten minutes after CPB the temperature change was -0.063 degrees C/min, and CO error was -0.31 +/- 0.36 (0.15 to -1.20) l/min. At 30 min the temperature change was -0.012 degrees C/min (not significant), and CO error was -0.13 +/- 0.14 l/min. Duration of CPB was 104 +/- 30 min, with rewarming for 44 +/- 13 min; the average minimum bladder temperature was 25.1 +/- 2.3 degrees C during cooling and 36.7 +/- 0.7 degrees C at the end of CPB. Under these conditions TDCO measurements within the first 10 min after CPB often underestimate the true CO.

Following cardiomyoplasty, the latissimus dorsi (LD) muscle contracts in response to both pacemaker and central nervous system (CNS) derived action potentials. To evaluate the effect of this dual stimulation on cardiac function, six goats (56 kg to 80 kg) underwent treadmill tests 4 to 5 months after surgery. Sonomicrometer crystals were used to measure changes in the left ventricular short-axis dimension (LVSAD) and regional length changes in the LD muscle. Pacing mediated stimulation resulted in 17 +/- 4% fractional shortening of the muscle pedicle lasting 400 +/- 17 msec. Walking at speeds of 0.8 and 2.4 km/hr produced muscle contractions of variable duration and intensity. LVSAD was decreased by both pacing (14%) and walking (up to 10%). Pacing in combination with walking caused erratic muscle contractions with concomitant changes in LVSAD suggestive of impaired ventricular filling. No arrhythmias were noted during exercise.

A simple, reliable method for the experimental production of pancreatic pseudocysts in dogs is described. With type A technique pseudocysts were produced in 100% of the animals with a 16% mortality.