The challenges of deep learning in artificial intelligence and autonomous actions in surgery: a literature review

Taher, Heba; Grasso, Vincent; Tawfik, Sherifa; Gumbs, Andrew A.

doi:10.20517/ais.2022.11

Cited by 38 publications

(35 citation statements)

References 24 publications

(79 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The economical results and long-term oncologic outcomes are not evaluated in our study, as adequate data are missing at present. With the technological advances of computer vision and artificial intelligence playing a role in the improvement of LPD, whether these theoretical advantages could translate into improved patient outcomes, especially for a more complex condition such as malignancy, needs further studies ( 82 , 83 ).…”

Section: Discussionmentioning

confidence: 99%

Laparoscopic versus open pancreaticoduodenectomy for pancreatic and periampullary tumor: A meta-analysis of randomized controlled trials and non-randomized comparative studies

Yan

Hua²,

Chen³

et al. 2023

Front. Oncol.

View full text Add to dashboard Cite

ObjectiveThis meta-analysis compares the perioperative outcomes of laparoscopic pancreaticoduodenectomy (LPD) to those of open pancreaticoduodenectomy (OPD) for pancreatic and periampullary tumors.BackgroundLPD has been increasingly applied in the treatment of pancreatic and periampullary tumors. However, the perioperative outcomes of LPD versus OPD are still controversial.MethodsPubMed, Web of Science, EMBASE, and the Cochrane Library were searched to identify randomized controlled trials (RCTs) and non-randomized comparative trials (NRCTs) comparing LPD versus OPD for pancreatic and periampullary tumors. The main outcomes were mortality, morbidity, serious complications, and hospital stay. The secondary outcomes were operative time, blood loss, transfusion, postoperative pancreatic fistula (POPF), postpancreatectomy hemorrhage (PPH), bile leak (BL), delayed gastric emptying (DGE), lymph nodes harvested, R0 resection, reoperation, and readmission. RCTs were evaluated by the Cochrane risk-of-bias tool. NRCTs were assessed using a modified tool from the Methodological Index for Non-randomized Studies. Data were pooled as odds ratio (OR) or mean difference (MD). This study was registered at PROSPERO (CRD42022338832).ResultsFour RCTs and 35 NRCTs concerning a total of 40,230 patients (4,262 LPD and 35,968 OPD) were included. Meta-analyses showed no significant differences in mortality (OR 0.91, p = 0.35), serious complications (OR 0.97, p = 0.74), POPF (OR 0.93, p = 0.29), PPH (OR 1.10, p = 0.42), BL (OR 1.28, p = 0.22), harvested lymph nodes (MD 0.66, p = 0.09), reoperation (OR 1.10, p = 0.41), and readmission (OR 0.95, p = 0.46) between LPD and OPD. Operative time was significantly longer for LPD (MD 85.59 min, p < 0.00001), whereas overall morbidity (OR 0.80, p < 0.00001), hospital stay (MD −2.32 days, p < 0.00001), blood loss (MD −173.84 ml, p < 0.00001), transfusion (OR 0.62, p = 0.0002), and DGE (OR 0.78, p = 0.002) were reduced for LPD. The R0 rate was higher for LPD (OR 1.25, p = 0.001).ConclusionsLPD is associated with non-inferior short-term surgical outcomes and oncologic adequacy compared to OPD when performed by experienced surgeons at large centers. LPD may result in reduced overall morbidity, blood loss, transfusion, and DGE, but longer operative time. Further RCTs should address the potential advantages of LPD over OPD.Systematic review registrationPROSPERO, identifier CRD42022338832.

show abstract

Section: Discussionmentioning

confidence: 99%

Laparoscopic versus open pancreaticoduodenectomy for pancreatic and periampullary tumor: A meta-analysis of randomized controlled trials and non-randomized comparative studies

Yan

Hua²,

Chen³

et al. 2023

Front. Oncol.

View full text Add to dashboard Cite

show abstract

“…SVM was considered a "black box" algorithm, which might create a challenging dilemma for those responsible for a potential decision-making medical error. Interpreting how the model reached its predictions was therefore necessary to enable researchers to assess whether their predictions were trustworthy (23). Accumulating evidence have identified several risk factors for CPSP, including presence and intensity of acute postoperative pain, type of surgery, age, body mass index (BMI) and psychosocial factors (24,25).…”

Section: Discussionmentioning

confidence: 99%

Identifying an optimal machine learning model generated circulating biomarker to predict chronic postoperative pain in patients undergoing hepatectomy

Hong

et al. 2023

Front. Surg.

View full text Add to dashboard Cite

Chronic postsurgical pain (CPSP) after hepatectomy is highly prevalent and challenging to treat. Several risk factors have been unmasked for CPSP after hepatectomy, such as acute postoperative pain. The current secondary analysis of a clinical study sought to extend previous research by investigating more clinical variables and inflammatory biomarkers as risk factors for CPSP after hepatectomy and sifting those strongly related to CPSP to build a reliable machine learning model to predict CPSP occurring. Participants included 91 adults undergoing hepatectomy who was followed 3 months postoperatively. Twenty-four hours after surgery, participants completed numerical rating scale (NRS) grading and blood sample collecting. Three months after surgery, participants also reported whether CPSP occurred through follow-up. The Random Forest and Support Vector Machine models were conducted to predict pain outcomes 3 months after surgery. The results showed that the SVM model had better performance in predicting CPSP which consists of acute postoperative pain (evaluated by NRS) and matrix metalloprotease 3 (MMP3) level. What's more, besides traditional cytokines, several novel inflammatory biomarkers like C-X-C motif chemokine ligand 10 (CXCL10) and MMP2 levels were found to be closely related to CPSP and a novel spectrum of inflammatory biomarkers was created. These findings demonstrate that the SVM model consisting of acute postoperative pain and MMP3 level predicts greater chronic pain intensity 3 months after hepatectomy and with this model, intervention administration before CPSP occurs may prevent or minimize CPSP intensity successfully.

show abstract

“…To avoid misdiagnosis in similar cases, attention should be paid to important symptoms, including refractory vomiting, vague abdominal pain, and weight loss (or inadequate weight gain in pregnant women). Perhaps in the near future, artificial intelligence modules in medical management systems can help us solve these problems ( 23 ).…”

Section: Discussionmentioning

confidence: 99%

Ileal adenocarcinoma in a young pregnant woman: A rare case report

Xiao

Cheng²,

Cao³

et al. 2023

Front. Oncol.

View full text Add to dashboard Cite

Primary intestinal malignancies account for only 1%–3% of all malignant gastrointestinal tumors. Adenocarcinomas are uncommonly located in the ileum. Ileal adenocarcinoma (IA) is rare and difficult to diagnose because of its location. IA is common in older men and rare in young pregnant women. A 23-year-old pregnant woman was hospitalized several times for repeated vomiting and abdominal pain. Her symptoms were relieved after symptomatic treatment. She exhibited no typical manifestations of intestinal obstruction, such as abdominal distension, difficulty passing gas and defecation. Unfortunately, she was misdiagnosed with acute gastroenteritis. On the second day after delivery, the patient stopped passing gas and computed tomography (CT) revealed an intestinal obstruction. She was treated as paralytic ileus. However, in view of failed conservative management, she was decided for an exploratory laparotomy. A malignant ileal tumor 5cm from the ileocecal valve was found incidentally and was surgically excised accompanied with End-to-side anastomosis of ileal and transverse colon. The operation lasted 195 minutes. Pathological examination revealed an IA. Pregnant woman who experience symptoms of intestinal obstruction should be alert to the possibility of malignancy in the small intestine. IA is an insidious tumor in pregnant women. An “IA triad” can be defined as refractory vomiting, vague abdominal pain, and weight loss (or inadequate weight gain in pregnant women). Pregnant women with an IA triad should undergo investigation with endoscopy or, if necessary, magnetic resonance imaging (MRI).

show abstract

The challenges of deep learning in artificial intelligence and autonomous actions in surgery: a literature review

Cited by 38 publications

References 24 publications

Laparoscopic versus open pancreaticoduodenectomy for pancreatic and periampullary tumor: A meta-analysis of randomized controlled trials and non-randomized comparative studies

Laparoscopic versus open pancreaticoduodenectomy for pancreatic and periampullary tumor: A meta-analysis of randomized controlled trials and non-randomized comparative studies

Identifying an optimal machine learning model generated circulating biomarker to predict chronic postoperative pain in patients undergoing hepatectomy

Ileal adenocarcinoma in a young pregnant woman: A rare case report

Contact Info

Product

Resources

About