Machine learning, deep learning and hernia surgery. Are we pushing the limits of abdominal core health? A qualitative systematic review.

INTRODUCTION: This systematic review aims to evaluate the use of machine learning and artificial intelligence in hernia surgery. METHODS: The PRISMA guidelines were followed throughout this systematic review. The ROBINS-I and Rob 2 tools were used to perform qualitative assessment of all studies included in this review. Recommendations were then summarized for the following pre-defined key items: protocol, research question, search strategy, study eligibility, data extraction, study design, risk of bias, publication bias, and statistical analysis. RESULTS: A total of 13 articles were ultimately included for this review, describing the use of machine learning and deep learning for hernia surgery. All studies were published from 2020 to 2023. Articles varied regarding the population studied, type of machine learning or Deep Learning Model (DLM) used, and hernia type. Of the thirteen included studies, all included either inguinal, ventral, or incisional hernias. Four studies evaluated recognition of surgical steps during inguinal hernia repair videos. Two studies predicted outcomes using image-based DMLs. Seven studies developed and validated deep learning algorithms to predict outcomes and identify factors associated with postoperative complications. CONCLUSION: The use of ML for abdominal wall reconstruction has been shown to be a promising tool for predicting outcomes and identifying factors that could lead to postoperative complications.

Robotic repair of post-oesophagectomy hiatal hernia.

Diaphragmatic hiatal hernia is a potential complication of oesophagectomy in cancer patients. Over the past decades, laparoscopy has become the preferred approach to repairing this condition due to the reduced morbidity, faster recovery and shorter hospital stay when compared with traditional open surgery. The development of robotic technology has added to the benefits of minimally invasive approaches, offering potential technical advantages and overcoming some limitations of traditional laparoscopic techniques when performing complex procedures. We present the first report of a robotic post-oesophagectomy hiatal hernia repair.

Critical view of robotic-assisted transverse abdominal release (r-TAR).

INTRODUCTION: Establishing straightforward and reproducible steps to describe the technique performed with the aid of the robotic system for complex hernia surgery is key for good outcomes. Even using the description of open surgery as a parameter for performing the robotic technique, it is important to stress the particularities of this access. To describe the steps to perform robotic-assisted TAR (r-TAR) in a standardized technique, with a critical and safe view of all the anatomical structures. DESCRIPTION OF THE TECHNIQUE: We defined 8 landmarks for the critical view of safety in r-TAR which include: (1) patient position, trocar and docking; (2) posterior rectus sheath mobilization; (3) transversus abdominis release (TAR)-Top-down technique; (4) transversus abdominis release (TAR)-bottom-up technique and mesh insertion; (5) contralateral trocar insertion and redocking, 6) posterior sheath closure; (7) final mesh positioning; and (8) anterior defect closure and drains. DISCUSSION: Complex hernia surgery using a robotic-assisted posterior component separation requires well-established steps so the procedure can be reproducible and achieve better results.

Robotic assisted eTEP ventral hernia repair: Brazilian early experience.

PURPOSE: The enhanced view totally extraperitoneal (eTEP) concept first applied to inguinal hernias has become an important tool in the armamentarium of ventral hernia repair. The purpose of this report is to date the first brazilian robotic-assisted eTEP ventral hernia repair case series. METHODS: A review of a prospectively maintained database was conducted in patients who underwent robotic-assisted enhanced view totally extraperitoneal (eTEP) for ventral hernia repair between June 2018 and January 2020. Patients demographics, preoperative hernia characteristics, intraoperative variables and postoperative outcomes were evaluated. RESULTS: Our review identified 74 patients submitted to the procedure. Thirty-one patients (41.8%) presented primary ventral hernias (PVHs) and 43 patients (58.2%) presented incisional hernias (IHs). Female patients were predominant in both groups PVHs and IHs with 17 (22.9%) and 22 (29.7%) respectively, with a total of 39 patients (52.7%). Mean BMI was 29.1 kg/m2 (range 21.3-48.0 kg/m2) with higher mean BMI rates of 30.3 kg/m2 in the IHs group (range 22-48 years). A lateral dock setup was utilized in 55 cases (74.4%), having the inferior and superior dock setup in 18 (24.3%) and 1 (1.3%) cases respectively. Mean console time was 148.3 (range 75-277 min) and 192.6 min (range 66-301 min) in the PVHs and IHs groups respectively. There were no intraoperative complications or conversions. Average length of stay was 1.5 days. Four patients were readmitted within 30 days of surgery. There were no reoperation or cases of intraparietal herniation in this cohort. No hernia recurrence was verified during the mean 230.7 days of follow up in both groups. CONCLUSION: We present the first brazilian series to-date of the robotic assisted eTEP approach for ventral hernia repairs. Although long term outcomes require further analysis, its feasibility and reproducibility in experienced surgeons hands are evident, with safe and acceptable early postoperative outcomes.

Ventral hernia secondary to giant omphalocele in a child: combined approach of botulinum toxin and preoperative progressive pneumoperitoneum.

Omphalocele is a congenital abdominal wall defect that occurs approximately 1 in 4000-6000 live births. The abdominal-visceral disproportion, large diameter of the defect, volume of liver in the sac along with high incidence of associated anomalies make the surgical management a real challenge. Currently, there are two strategies for managing giant omphaloceles, staged surgical closure and nonoperative delayed closure. The combined treatment with PPP and BoNT/A injection has recently been described in adults. There is strong evidence on safety and efficacy of the use of BoNT/As in other areas of pediatrics and no recent reports of PPP use in children. Also, there are no data available about the combination of both techniques in pediatric population. The purpose of this manuscript is to report a case of a 7-year-old female child that was referred to our institution with a large ventral hernia secondary to omphalocele. We opted for a combined approach with BoNT/A injection and PPP before the definitive surgery. The surgical result was great with midline closure with no tension and no need for prosthetic substitution or component separation needed. To our knowledge, this is the first case report of BoNT/A injection and PPP for large ventral hernias in children. BoNT/A application was safe and the PPP technique was also proved to be applicable on children. We believe that the combination of BoNT/A and PPP presented to be a safe approach with an excellent result, particularly for not needing abdominal wall prosthetic substitution.

A laparoscopy-based method for BRDF estimation from in vivo human liver.

While improved visual realism is known to enhance training effectiveness in virtual surgery simulators, the advances on realistic rendering for these simulators is slower than similar simulations for man-made scenes. One of the main reasons for this is that in vivo data is hard to gather and process. In this paper, we propose the analysis of videolaparoscopy data to compute the Bidirectional Reflectance Distribution Function (BRDF) of living organs as an input to physically based rendering algorithms. From the interplay between light and organic matter recorded in video images, we propose the definition of a process capable of establishing the BRDF for inside-the-body organic surfaces. We present a case study around the liver with patient-specific rendering under global illumination. Results show that despite the limited range of motion allowed within the body, the computed BRDF presents a high-coverage of the sampled regions and produces plausible renderings.

Facilitating gastrotomy closure during natural-orifice transluminal endoscopic surgery using tissue anchors.

BACKGROUND AND STUDY AIMS: Reliable and secure closure of the gastrotomy after natural-orifice transluminal endoscopic surgery (NOTES) remains a critical step for widespread acceptance and use of this mode of surgery. We describe a novel method for gastrotomy closure using endoscopic tissue anchors. METHODS: A standard upper endoscopy and wire placement as used for percutaneous endoscopic gastrostomy placement was performed in five pigs. Prior to gastrotomy, four tissue anchors were placed in four quadrants (1 cm away from the wire). A 12-mm gastrotomy was created endoscopically using a combination of needle-knife and balloon dilation. After transgastric peritoneoscopy, the sutures were approximated using a device knotting element. One additional pair of sutures was placed after evaluation of the gastric closure. The animals underwent in vivo contrast fluoroscopy, methylene blue instillation, and bursting pressure studies for assessment of the closure site. RESULTS: All animals studied showed complete sealing of the gastrotomy site without evidence of leak on fluoroscopic imaging or at final postmortem intragastric methylene blue instillation. Improved insufflation ability following gastrotomy was also noted using this technique, which enhanced overall visualization during the closure. CONCLUSION: Positioning tissue anchors prior to creating a NOTES gastrotomy was a feasible and reliable method to perform gastric closure. Follow-up survival studies will be warranted to support these preliminary findings.

Influence of surgically induced gastric and gastroduodenal content reflux on esophageal carcinogenesis--experimental model in Wistar female rats.

Studies in human beings and animals have shown that esophageal exposure to duodenal and gastric contents may be important for the development of Barrett's esophagus and its complications, including adenocarcinoma and epidermoid carcinoma. Diethylnitrosamine (DEN) is a carcinogen that stimulates the development of epidermoid carcinoma in the esophagus of mice. The aim of this study was to evaluate the effect of gastroduodenal and gastric content reflux on induction of esophageal carcinogenesis. Gastroesophageal reflux (GER) and gastroduodenoesophageal reflux (GDER) were produced by cardioplasty and esophagoduodenostomy. The chosen carcinogen was DEN, diluted in drinking water, given 3 days a week for 20 consecutive weeks. One hundred Wistar female rats were divided into six groups, as follows: group 1 (18 rats), cardioplasty without DEN; group 2 (18 rats), cardioplasty with DEN; group 3 (10 rats), only water; group 4 (17 rats), cardioplasty with DEN; group 5 (17 rats), esophagoduodenostomy with DEN; group 6 (20 rats), only DEN. GER in isolation induced papillomatosis or ulceration in 22.2% of rats and, when associated with DEN, induced papillomatosis in 61.1% of rats. GDER in isolation induced marked esophagitis in 61.1% of rats, Barrett's esophagus in 16.7% and esophageal adenocarcinoma in 16.7%; when associated with DEN, 23.5% of rats presented marked esophagitis, papillomatosis or ulceration, whereas 76.5% had esophageal carcinoma, with 70.6% epidermoid carcinoma and 5.9% adenocarcinoma. Rats treated with water alone did not show histologic abnormalities of the esophageal mucosa. Rats treated with DEN alone developed papillomas in 50.0% of the cases and remained histologically unchanged in 50.0%. There was no development of low- or high-grade dysplasia in any group. The conclusions are that (1) GDER is significantly more deleterious to esophageal mucosa than GER; (2) in this study, GER did not present carcinogenic potential in relation to the esophagus; (3) GDER in isolation is an esophageal carcinogen, producing Barrett's esophagus and esophageal adenocarcinoma; (4) esophageal oncogenesis caused by GDER is potentiated by DEN, inducing esophageal epidermoid carcinoma; (5) in this study, DEN in isolation did not generate tumors in the esophagus of rats.