運用紅外線熱像儀偵測乳牛蹄病

Abstract

跛足、乳房炎與繁殖障礙為現今乳牛三大淘汰原因。即時偵測牛群中罹患蹄病之乳牛並給予治療是相當重要的，除了有助於減緩或反轉蹄病之進程之外，還能避免蹄病衍生的負面影響，並延長乳牛之使用年限。過去，我們透過肉眼觀察乳牛行走時的步態與姿勢，篩選跛足乳牛進行治療，但因此種方式費時、費力與過於主觀，尋求替代方案勢在必行。 現今，隨著科技進步，越來越多新興的電子產品問世，其中紅外線熱像儀因能測量物體表面溫度之變化，而被應用於檢測發炎反應，並作為乳牛蹄病診斷工具之一。然而，環境中的溫度、濕度、風速與太陽輻射皆會影響紅外線熱像儀之測量結果，因此研究學者對於在高溫環境中使用紅外線熱像儀偵測蹄病的可行性存有疑慮。故本試驗之目的有三，其一在探討亞熱帶氣候下藉由紅外線熱像儀偵測蹄病之可行性，其二為研究不同拍照方向是否會影響蹄部表面溫度之測量結果，以及探討整合乳牛行為數據與蹄部表面最高溫作為蹄病診斷工具之可行性。 結果顯示雖然環境溫度與蹄部表面溫度呈現正相關，但在高溫環境下，透過紅外線熱像儀測量到的患病蹄之溫度仍然顯著高於健康蹄。此外，患病蹄在不同拍照方向下的蹄部表面最高溫皆較健康蹄來的高。整合蹄部表面最高溫與乳牛活動量數據之蹄病判讀結果亦與臨床診斷達到75%之一致性。 總結而言，在亞熱帶高溫氣候下紅外線熱像儀具有偵測蹄病之潛力。三個拍照方向皆能藉由蹄部表面最高溫區分健康與患病蹄，可提供不同動線的牧場紅外線熱像儀架設之參考依據。畜牧業未來牧場管理趨勢為整合多個不同類型的自動化監測資料之結果，並提供牧場管理者即時警訊與解決方案建議，而在本試驗中結合紅外線熱像儀測量之蹄部溫度與乳牛活動量數據，可供作篩選牛群中蹄病牛，為日後相關研究開出先例，並藉由行為數據之輔助，使環境溫度對紅外線熱像儀判讀之干擾得以減少。

Lameness, mastitis and reproductive disorder are the main culling causes in dairy cows. It is important to find and treat the cows with hoof lesions in the herd as early as possible. It can not only slow down or reverse the progress of hoof lesions, but also avoid derived negative effect and consequence. Moreover, it will prolong dairy cows’ longevity. In the past, people identified lame cows by observing their walking gaits and posture. This method is time-consuming, laborious, and rather subjective. Therefore, it is imperative to find an alternative way to detect hoof disease in dairy cattle. Nowadays, with the advancement of technology, more and more electronic devices come out. Infrared thermography (IRT) is used as an inflammation detection tool, because it can measure the temperature change on the surface of the objects, and become one of the hoof lesions’ detecting devices for dairy cows. However, ambient temperature, humidity, wind speed, and radiation may affect the measurement of the IRT. Thence, researchers have a doubt about using IRT as a hoof lesion detection tool under high environmental temperature conditions. There were three purposes in this research proposal. First, to investigate the feasibility of detecting hoof lesion through IRT under subtropical climate conditions. Second, to examine the effect of different photo shooting angles on IRT measurement. Finally, to test the possibility to integrate monitoring of maximum hoof surface temperature by IRT and continuous data collection by automated behavior accelerometer to establish a new hoof lesion detection model. It is found that ambient temperature had positive correlation with hoof surface temperature, but the temperature of the hooves with lesion was still significantly higher than sound hooves in hot environments. Furthermore, hooves with lesion their max surface temperature was significantly higher than sound hooves in all three different photo shooting directions. The reliability of integrated maximum hoof surface temperature and activity accelerometer data as a hoof lesions’ detecting tool reached 75% precision based on confirmation by clinical diagnosis. In conclusion, IRT has the potential as a hoof lesion detection tool under subtropical climate conditions. All three different photo shooting directions of IRT could identify the sound hooves and hooves with lesion through max surface temperature measurement, and provided the basis for on-site IRT device installation. The future trend of animal agriculture is to integrate multiple types of automatic collected data, and provide real-time warning and recommend solutions to the farm managers. This is the first study that integrates max surface temperature measurement from IRT and accelerometer activity data to identify the lesion cows in the herd. With the aid of behavioral accelerometer data, the interference of ambient temperature on the IRT is reduced.