Feasibility and basic acoustic characteristics of intelligent long-term bowel sound analysis in term neonates
ObjectiveBowel dysfunction continues to be a serious issue in neonates. Traditional auscultation of bowel sounds as a diagnostic tool in neonatal gastrointestinal problems is limited by skill and inability to document and reassess. Consequently, in order to objectively and noninvasively examine the...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-11-01
|
| Series: | Frontiers in Pediatrics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fped.2022.1000395/full |
| _version_ | 1811334970575159296 |
|---|---|
| author | Ping Zhou Meiling Lu Ping Chen Danlei Wang Zhenchao Jin Lian Zhang |
| author_facet | Ping Zhou Meiling Lu Ping Chen Danlei Wang Zhenchao Jin Lian Zhang |
| author_sort | Ping Zhou |
| collection | DOAJ |
| description | ObjectiveBowel dysfunction continues to be a serious issue in neonates. Traditional auscultation of bowel sounds as a diagnostic tool in neonatal gastrointestinal problems is limited by skill and inability to document and reassess. Consequently, in order to objectively and noninvasively examine the viability of continuous assessment of bowel sounds, we utilized an acoustic recording and analysis system to capture bowel sounds and extract acoustic features in term neonates.MethodsFrom May 1, 2020 to September 30, 2020, 82 neonates who were hospitalized because of hyperbilirubinemia were included. For 20 h, a convolutional neural network-based acoustic recorder that offers real-time, wireless, continuous auscultation was employed to track the bowel sounds of these neonates.Results(1) Usable data on five acoustic parameters of bowel sound was recorded for 68 neonates, and the median values were as follows: The rate was 25.80 times/min [interquartile range (IQR): 15.63–36.20]; the duration was 8.00 s/min (IQR: 4.2–13.20); the amplitude was 0.46 (IQR: 0.27–0.68); the frequency was 944.05 Hz (IQR: 848.78–1,034.90); and the interval time was 2.12 s (IQR: 1.3–3.5). (2) In comparison to the parameters of the bowel sounds recorded from the right lower abdomen in 68 infants, the acoustic parameters of the 10 out of 68 infants from chest controls and blank controls were considerably different. (3) The 50%–75% breast milk intake group had the highest rate, the longest duration, and the highest amplitude of bowel sounds, while the >75% breast milk intake group had the highest frequency of bowel sounds. (4) Compared with neonates without hyperbilirubinemia, there was no significant difference in the five parameters of bowel sounds in hyperbilirubinemia infants; nor was there a significant effect of phototherapy and non-phototherapy status on the parameters of bowel sounds during bowel sound monitoring in hyperbilirubinemia patients. (5) A mild transient skin rash appeared on the skin of three infants. No other adverse events occurred.ConclusionThe acoustic recording and analysis system appears useful for monitoring bowel sounds using a continuous, invasive, and real-time approach. Neonatal bowel sounds are affected by various feeding types rather than hyperbilirubinemia and phototherapy. Potential influencing factors and the significance of their application in neonatal intestinal-related disorders require further research. |
| first_indexed | 2024-04-13T17:16:12Z |
| format | Article |
| id | doaj.art-effb254df50947a8ad3a12cb037416ac |
| institution | Directory Open Access Journal |
| issn | 2296-2360 |
| language | English |
| last_indexed | 2024-04-13T17:16:12Z |
| publishDate | 2022-11-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Pediatrics |
| spelling | doaj.art-effb254df50947a8ad3a12cb037416ac2022-12-22T02:38:07ZengFrontiers Media S.A.Frontiers in Pediatrics2296-23602022-11-011010.3389/fped.2022.10003951000395Feasibility and basic acoustic characteristics of intelligent long-term bowel sound analysis in term neonatesPing Zhou0Meiling Lu1Ping Chen2Danlei Wang3Zhenchao Jin4Lian Zhang5Department of Neonatology, Baoan Women's and Children's Hospital, Shenzhen, ChinaDepartment of Neonatology, Baoan Women's and Children's Hospital, Shenzhen, ChinaResearch and Development Department, Linkwah Integrated Circuit Institute, Nanjing, ChinaResearch and Development Department, Linkwah Integrated Circuit Institute, Nanjing, ChinaDepartment of Neonatology, Baoan Women's and Children's Hospital, Shenzhen, ChinaDepartment of Neonatology, Baoan Women's and Children's Hospital, Shenzhen, ChinaObjectiveBowel dysfunction continues to be a serious issue in neonates. Traditional auscultation of bowel sounds as a diagnostic tool in neonatal gastrointestinal problems is limited by skill and inability to document and reassess. Consequently, in order to objectively and noninvasively examine the viability of continuous assessment of bowel sounds, we utilized an acoustic recording and analysis system to capture bowel sounds and extract acoustic features in term neonates.MethodsFrom May 1, 2020 to September 30, 2020, 82 neonates who were hospitalized because of hyperbilirubinemia were included. For 20 h, a convolutional neural network-based acoustic recorder that offers real-time, wireless, continuous auscultation was employed to track the bowel sounds of these neonates.Results(1) Usable data on five acoustic parameters of bowel sound was recorded for 68 neonates, and the median values were as follows: The rate was 25.80 times/min [interquartile range (IQR): 15.63–36.20]; the duration was 8.00 s/min (IQR: 4.2–13.20); the amplitude was 0.46 (IQR: 0.27–0.68); the frequency was 944.05 Hz (IQR: 848.78–1,034.90); and the interval time was 2.12 s (IQR: 1.3–3.5). (2) In comparison to the parameters of the bowel sounds recorded from the right lower abdomen in 68 infants, the acoustic parameters of the 10 out of 68 infants from chest controls and blank controls were considerably different. (3) The 50%–75% breast milk intake group had the highest rate, the longest duration, and the highest amplitude of bowel sounds, while the >75% breast milk intake group had the highest frequency of bowel sounds. (4) Compared with neonates without hyperbilirubinemia, there was no significant difference in the five parameters of bowel sounds in hyperbilirubinemia infants; nor was there a significant effect of phototherapy and non-phototherapy status on the parameters of bowel sounds during bowel sound monitoring in hyperbilirubinemia patients. (5) A mild transient skin rash appeared on the skin of three infants. No other adverse events occurred.ConclusionThe acoustic recording and analysis system appears useful for monitoring bowel sounds using a continuous, invasive, and real-time approach. Neonatal bowel sounds are affected by various feeding types rather than hyperbilirubinemia and phototherapy. Potential influencing factors and the significance of their application in neonatal intestinal-related disorders require further research.https://www.frontiersin.org/articles/10.3389/fped.2022.1000395/fullneonatebowel soundsconvolutional neural networksartificial intelligenceacoustic parameter |
| spellingShingle | Ping Zhou Meiling Lu Ping Chen Danlei Wang Zhenchao Jin Lian Zhang Feasibility and basic acoustic characteristics of intelligent long-term bowel sound analysis in term neonates Frontiers in Pediatrics neonate bowel sounds convolutional neural networks artificial intelligence acoustic parameter |
| title | Feasibility and basic acoustic characteristics of intelligent long-term bowel sound analysis in term neonates |
| title_full | Feasibility and basic acoustic characteristics of intelligent long-term bowel sound analysis in term neonates |
| title_fullStr | Feasibility and basic acoustic characteristics of intelligent long-term bowel sound analysis in term neonates |
| title_full_unstemmed | Feasibility and basic acoustic characteristics of intelligent long-term bowel sound analysis in term neonates |
| title_short | Feasibility and basic acoustic characteristics of intelligent long-term bowel sound analysis in term neonates |
| title_sort | feasibility and basic acoustic characteristics of intelligent long term bowel sound analysis in term neonates |
| topic | neonate bowel sounds convolutional neural networks artificial intelligence acoustic parameter |
| url | https://www.frontiersin.org/articles/10.3389/fped.2022.1000395/full |
| work_keys_str_mv | AT pingzhou feasibilityandbasicacousticcharacteristicsofintelligentlongtermbowelsoundanalysisintermneonates AT meilinglu feasibilityandbasicacousticcharacteristicsofintelligentlongtermbowelsoundanalysisintermneonates AT pingchen feasibilityandbasicacousticcharacteristicsofintelligentlongtermbowelsoundanalysisintermneonates AT danleiwang feasibilityandbasicacousticcharacteristicsofintelligentlongtermbowelsoundanalysisintermneonates AT zhenchaojin feasibilityandbasicacousticcharacteristicsofintelligentlongtermbowelsoundanalysisintermneonates AT lianzhang feasibilityandbasicacousticcharacteristicsofintelligentlongtermbowelsoundanalysisintermneonates |